US20170085863A1 - Method of converting 2d video to 3d video using machine learning - Google Patents
Method of converting 2d video to 3d video using machine learning Download PDFInfo
- Publication number
- US20170085863A1 US20170085863A1 US14/967,939 US201514967939A US2017085863A1 US 20170085863 A1 US20170085863 A1 US 20170085863A1 US 201514967939 A US201514967939 A US 201514967939A US 2017085863 A1 US2017085863 A1 US 2017085863A1
- Authority
- US
- United States
- Prior art keywords
- depth
- frames
- model
- machine learning
- conversion
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N13/00—Stereoscopic video systems; Multi-view video systems; Details thereof
- H04N13/10—Processing, recording or transmission of stereoscopic or multi-view image signals
- H04N13/106—Processing image signals
- H04N13/122—Improving the 3D impression of stereoscopic images by modifying image signal contents, e.g. by filtering or adding monoscopic depth cues
-
- H04N13/0264—
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/50—Depth or shape recovery
- G06T7/55—Depth or shape recovery from multiple images
- G06T7/557—Depth or shape recovery from multiple images from light fields, e.g. from plenoptic cameras
-
- H04N13/0018—
-
- H04N13/0022—
-
- H04N13/0275—
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N13/00—Stereoscopic video systems; Multi-view video systems; Details thereof
- H04N13/10—Processing, recording or transmission of stereoscopic or multi-view image signals
- H04N13/106—Processing image signals
- H04N13/128—Adjusting depth or disparity
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N13/00—Stereoscopic video systems; Multi-view video systems; Details thereof
- H04N13/20—Image signal generators
- H04N13/261—Image signal generators with monoscopic-to-stereoscopic image conversion
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20081—Training; Learning
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N13/00—Stereoscopic video systems; Multi-view video systems; Details thereof
- H04N2013/0074—Stereoscopic image analysis
- H04N2013/0081—Depth or disparity estimation from stereoscopic image signals
Landscapes
- Engineering & Computer Science (AREA)
- Multimedia (AREA)
- Signal Processing (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Image Processing (AREA)
- Processing Or Creating Images (AREA)
Abstract
Description
- This application is a continuation in part of U.S. Utility patent application Ser. No. 14/857,704, filed 17 Sep. 2015, the specification of which is hereby incorporated herein by reference.
- Field of the Invention
- One or more embodiments of the invention are related to the field of image processing. More particularly, but not by way of limitation, one or more embodiments of the invention enable a method of converting 2D video to 3D video using machine learning. Embodiments of the invention train a machine learning system to perform one or more 2D to 3D conversion steps. The machine learning system is trained on a training set that includes 2D to 3D conversion examples; it derives generalized 2D to 3D transformation functions from this training set. Embodiments of the invention may also obtain 3D models of objects, such as characters, and process the frames of a video to locate and orient these models in the frames. Depth maps and stereoscopic 3D video may then be generated from the 3D models. Embodiments of the invention may convert 3D scanner data to a set of rotated planes associated with masked areas of the data, thus forming a 3D object model. This enables the planes to be manipulated independently or as part of a group, and eliminates many problems associated with importing external 3D scanner data including minimization of errors that frequently exist in external 3D scanner data.
- Description of the Related Art
- Two-dimensional images contain no depth information and hence appear the same to an observer's left and right eye. Two-dimensional images include paper photographs or images displayed on a standard computer monitor. Two-dimensional images however may include shading and lighting that provide the observer a sense of depth for portions of the image, however, this is not considered a three-dimensional view of an image. Three-dimensional images on the other hand include image information that differs for each eye of the observer. Three-dimensional images may be displayed in an encoded format and projected onto a two-dimensional display. This enables three-dimensional or stereoscopic viewing for example with anaglyph glasses or polarized glasses. Other displays may provide different information based on the orientation with respect to the display, e.g., autostereoscopic displays that do not require special glasses for viewing three-dimensional images on a flat two-dimensional display. An example of such as display is a lenticular display. Alternatively, two images that are shown alternately to the left and right eyes may be viewed with shutter glasses. Regardless of the type of technology involved, conversion of two-dimensional images to stereoscopic images requires the addition of depth information to the two-dimensional input image.
- Current solutions for conversion of two-dimensional images to stereoscopic images generally require large amounts of manual labor for highly accurate results. These manual masking systems generally operate by accepting manually created masks in order to define areas or regions in the image that have different depths and which generally represent different human observable objects. Depth information is then accepted by the system as input from artists for example, which results in nearer objects being shifted relatively further horizontally to produce left and right eye viewpoints or images, or Red/Blue anaglyph single image encodings, either of which may be utilized for stereoscopic viewing. By shifting objects in the foreground, hidden or background information may be exposed. If the missing image data is not shown in any other images in a scene, then the “gap” must be filled with some type of image data to cover the artifact. If the hidden image data does not exist in any other image in a scene, then this prohibits borrowing of pixels from the areas in other images that do contain the missing information. Various algorithms exist for filling gaps, which are also known as occlusion filling algorithms, to minimize the missing information with varying success. Generally, the depth artist gains visual clues from the image and applies depth to masks using artistic input.
- The 2D to 3D conversion processes described above require large amount of manual labor. There are no known systems that automate the conversion process. However, because some organizations have performed hundreds or thousands of 2D to 3D conversions, there is a considerable database of conversion examples. In principle, machine learning techniques can be applied to develop generalized 2D to 3D conversion methods from such a historical database of conversion examples. Machine learning techniques are known in the art, but they have not been applied to 2D to 3D conversion. There are no known systems that apply machine learning techniques to develop 2D to 3D conversion methods using a database of conversion examples.
- For at least the limitations described above there is a need for a method to convert 2D video to 3D video using machine learning.
- One or more embodiments described in the specification are related to a method of converting of 2D video to 3D video using 3D object models that provides increased artistic and technical flexibility and rapid conversion of movies for stereoscopic viewing. Embodiments of the invention convert a large set of highly granular depth information inherent in a depth map associated with a two-dimensional image to a smaller set of rotated planes associated with masked areas in the image. This enables the planes to be manipulated independently or as part of a group and eliminates many problems associated with importing external depth maps including minimization of errors that frequently exist in external depth maps
- Embodiments of the invention may utilize any type of depth map including Z-Depth associated with images that are generated through rendering from a Computer Generated Imagery or CGI application such as MAYA® or HOUDINI®, depth maps obtained after conversion of a disparity map from a stereoscopic pair of images to a Z-Depth, Z-Depth extraction from of a light-field image, time-of-flight imaging systems, LIDAR, or any other type of depth map associated with a two-dimensional image.
- Embodiments of the invention include a number of inherent advantages over simply using the Z-Depths as is currently performed in automated or semi-automated 2D to 3D conversion processes.
- For example, embodiments of the invention transform the large set of depth map depths or Z-Depth into a manageable number of parts. Thus, the system enables an artist to manipulate individual or groups of parts for artistic purposes, as opposed to pixel-by-pixel editing. So, for example, an artist may independently adjust the angle, and hence depth of a robot's arm so the resulting stereoscopic image appears to reach out of the screen.
- In addition, by transforming the Z-Depth into a manageable number of parts, the system enables an artist to group these parts and apply separate RGB image layers to these groups. This enables more efficient occlusion filling in the 2D to 3D conversion workflow.
- Furthermore, embodiments of the invention mold depth data to eliminate depth errors by transforming large numbers of depth values to smaller number of plane rotations. In one embodiment, the system may calculate the normal and position for a specific region, for example to form an average, rotation value associated with a plane that represents a large group of depth values, some of which may be erroneous. Hence, issues associated with imperfect depth map data are often averaged out, or otherwise eliminated. In some extreme cases of noisy depth data, these issues may not be fully resolved, however, embodiments of the invention reduce the problem to a manageable number of editable parts, and enable the issues to be rapidly and easily corrected automatically or by accepting inputs from an artist. One or more embodiments of the invention may utilize a normal vector algorithm. Other algorithms may be utilized alone or in combination with the normal vector method to achieve similar or advantageous results. For example, embodiments of the invention may treat each pixel as a point in space, e.g., wherein X and Y represent the position of the pixel and Z represents the Z-Depth value of that pixel, and isolate only the points within the defined region, and calculate the “best-fit” plane for that group of points, and/or a normal vector representation of the plane. The normal vector in this embodiment is orthogonal to the plane and may be encoded into separate RGB channels in order to provide a viewable representation of the planar angles with respect to the optical display. Embodiments of the invention may utilize any type of plane fitting algorithm including, but not limited to, regression plane, orthogonal distance regression plane, etc. Embodiments of the invention may utilize any type of filtering as part of the transformation processing including but not limited to dilation and erosion.
- One or more embodiments of the invention implement a method on a computer for example wherein the method includes obtaining an external depth map associated with a two-dimensional image, obtaining at least one mask associated with at least one area within the two-dimensional image, calculating a fit or best fit for a plane using a computer based on depth associated with the at least one area associated with each of the at least one mask, applying depth associated with the plane having the fit to the at least one area to shift pixels in the two-dimensional image horizontally to produce a stereoscopic image or stereoscopic image pair.
- Embodiments of the method may also include obtaining of the external depth map associated with a two-dimensional image by obtaining a disparity map, or a depth map of lower resolution than the two-dimensional image from a pair of witness cameras, or a depth map from time-of-flight system, or a depth map from a triangulation system.
- Embodiments of the invention may also include obtaining at least one mask associated with at least one area within the two-dimensional image by automatically generating the at least one mask comprising the at least one area wherein the at least one area is over a predefined size and within a predefined depth range, or automatically generating the at least one mask comprising the at least one area wherein the at least one area comprises a boundary having a difference in luminance values over a predefined threshold, or both methods of size, depth range and boundary or any combination thereof.
- Embodiments of the invention may also include calculating the best fit for a plane using a computer based on depth associated with the at least one area associated with each of the at least one mask by calculating a normal vector for the plane, or a regression fit for the plane, or an orthogonal distance regression fit for the plane, or in any other known manner regarding fitting a plane to particulars points in three-dimensional space.
- Embodiments of the invention generally also include applying depth associated with the plane having the best fit to the at least one area to shift pixels in the two-dimensional image horizontally to produce a stereoscopic image or stereoscopic image pair.
- Embodiments may also include grouping two or more of the planes in order to provide a piecewise masked surface area. The grouping may include a link of a predefined minimum and maximum distance, which enables moving one plane to move other grouped planes if the maximum values are hit. The minimum values may be zero or negative to allow to precise joining of planes or slight overlap for example. In one or more embodiments, the grouping may include a link having a spring constant, this enables the movement of planes relative to one another to self align with respect to the other planes to minimize the overall spring force on three or more of the corner points of the plane. Alternatively or in combination, embodiments of the invention may include altering automatically any combination of position, orientation, shape, depth or curve of said plane in order to fit edges or corners of the plane with another plane. This enables a plane to be positioned in three-dimensional space, rotated in three-dimensions, reshaped by moving a corner point, warped in effect by adding depth or a curve to the plane, for example to add depth to the plane itself to match the underlying image data. Embodiments of the invention may also include accepting an input to alter any combination of position, orientation, shape, depth or curve of the plane, for example to artistically fit the underlying image data, correct errors or artifacts from the automated fitting process for touch up, etc.
- One or more embodiments of the invention use one or more 3D object models to convert all or portions of a 2D video to a 3D video. A 3D object model may be obtained from any source and in any format for any object or objects that appear in one or more frames of a 2D video. For a scene of the 2D video containing the object, a 3D space may be defined for example with coordinate axes for the location of the scene. A 3D object model may then be positioned and oriented in this 3D space for each frame in which the corresponding object appears. From this positioned and oriented 3D object model, a depth map for the object in the frame may be generated. Embodiments may augment the depth map with depth information for other elements in the frame. A stereoscopic image pair may then be generated from the depth map and from the 2D frame using techniques known in the art for 2D to 3D conversion.
- One or more embodiments may obtain 3D scanner data for an object, and convert this scanner data into a 3D model for the object. 3D data may be obtained using any 3D scanning technology, including for example laser systems using time-of-flight or triangulation, or systems using structured light fields. 3D data may be obtained from stereo cameras or systems with witness cameras. Any technique for obtaining 3D data describing an object is in keeping with the spirit of the invention.
- Conversion of 3D data to a 3D object model may for example include retopologizing the 3D data to a lower polygon count model or to a model with parameterized surfaces. One or more embodiments may use retopologizing techniques to reduce model complexity. Conversion may for example include fitting one or more planes to the 3D data and generating a 3D object model from these planes. Any of the techniques described above for fitting and adjusting planes or other surfaces may be used in one or more embodiments to generate a 3D object model. For example, one or more planes may be defined using regression fit, orthogonal distance regression fit, or calculation of normal vectors to the planes. One or more embodiments may combine multiple planes into groups of planes with links between the planes; links may for example define constraints between the edges or corners of the planes. One or more embodiments may employ link constraints that define maximum or minimum values for distances or angles. One or more embodiments may employ link constraints that use spring constants for the links and that for example minimize a potential energy of the model using the spring constants. One or more embodiments may define masks for selected areas of the model, using for example areas of predefined size or predefined depth, or areas with boundaries determined by changes in luminance or other image features over predefined thresholds.
- One or more embodiments may adjust the planes or surfaces fit to 3D data, for example by altering one or more of position, orientation, shape, depth or curve of the more planes in order to fit edges or corners of said the planes with one another. Adjustments may be made manually or may be made automatically using a computer. The adjusted planes or surfaces may be used to generate a 3D object model.
- One or more embodiments may locate and orient a 3D model in a frame using one or more model features. Features on the 3D object model may be aligned with pixel locations in the frame, and the 3D object model position and orientation in the 3D space of the scene may be calculated using techniques known in the art to minimize errors between observed pixel coordinates and the projection of the 3D feature points onto the image plane. One or more embodiments may align 3D object models with images in a frame in one or more key frames, and automatically determine the alignment in non-key frames using automated feature tracking.
- In one or more embodiments one or more objects may be non-rigid in that they comprise multiple parts that may move relative to one another. One or more embodiments may generate or obtain rigged 3D object models with degrees of freedom between the parts of the models. Locating and orienting these rigged models in frames may include determining values for the degrees of freedom of the rigged models. These values may be determined manually, automatically, or using a combination of manual and automated techniques. One or more embodiments may determine the location and orientation of each part of a rigged 3D object model by aligning features of the parts with pixels in an image of the object in a frame. Values for degrees of freedom may be interpolated automatically between key frames. One or more embodiments may manually locate and orient an object and define values for degrees of freedom in key frames, and may use feature tracking to automatically determine the object location and orientation and the values of the degrees of freedom in non-key frames.
- One or more embodiments of the invention use machine learning for conversion of 2D video to 3D video. A machine learning system may be trained on a training set comprising conversion examples. A conversion example describes the conversion of a 2D scene to 3D. It may for example contain a 3D conversion dataset that includes the inputs and outputs from one or more 2D to 3D conversion steps. 2D to 3D conversion steps may include, for example locating and identifying one or more objects in one or more 2D frames, generating object masks for these objects, generating depth models for these objects, generating a stereoscopic image pair for each 2D frame, and filling gaps in the image pair with pixel values for the missing pixels. A machine learning system may learn to perform any or all of these steps using the training set to develop the machine learning algorithms and functions. The machine learning system may then be used to perform 2D to 3D conversion on a new 2D video. In one or more embodiments some 2D to 3D conversion steps may be performed automatically by a machine learning system; other steps may be performed, completed, or modified by an operator.
- One or more embodiments may use machine learning to perform object masking. For example, a 3D conversion dataset associated with a conversion example in the training set may include a masking input and a masking output. The masking input may include, for example, an object identity and a location of one or more feature points of the object in one or more frames. The masking output may include, for example, a path defining a boundary of a masked region for the object. The path may comprise, for example, one or more segments, each of which may be a curve defined by one or more control points. The machine learning system may learn a function that maps these masking inputs to these masking outputs.
- One or more embodiments may use machine learning to generate an object depth model. For example, a 3D conversion dataset associated with a conversion example in the training set may include an object depth model input, and an object depth model output. The object depth model input may include, for example, an object mask. The corresponding output may include any information that assigns depth to one or more points within the mask. For example, an object depth model output may include a 3D point cloud giving 3D coordinates for points within the mask. An object depth model output may include a geometric model comprising regions within the mask, and associating each region with a planar or curved surface in 3D space. The machine learning system may learn a function that maps object depth model inputs, such as a mask, into the corresponding object depth model outputs.
- One or more embodiments may use machine learning to perform gap filling. One method of gap filling used by one or more embodiments is to generate a clean plate frame from one or more 2D frames, and to use pixels from the clean plate frame to fill missing pixels. A 3D conversion dataset associated with a conversion example in the training set may include a clean plate input and a clean plate output. The clean plate input may include for example a series of 2D frames; collectively these frames may contain all of the clean plate pixels, but the machine learning system needs to learn how to extract them from the individual frames. The clean plate output may include the clean plate frame. The machine learning system may learn a function that maps 2D frames into the corresponding clean plate frame.
- One or more embodiments may use machine learning to perform all of the steps of object masking, object depth modeling, and clean plate generation for gap filling.
- One or more embodiments may use a 3D object model to perform object masking, object depth modeling, or both. A 3D object model may be obtained, for example, from 3D scanner data. One or more embodiments may combine machine learning methods with the use of 3D object models to perform 2D to 3D conversion.
- The patent or application file contains at least one drawing executed in color. Copies of this patent or patent application publication with color drawing(s) will be provided by the Office upon request and payment of the necessary fee.
- The above and other aspects, features and advantages of the invention will be more apparent from the following more particular description thereof, presented in conjunction with the following drawings wherein:
-
FIG. 1 illustrates an exemplary overall system architecture for one or more embodiments of the invention. -
FIG. 2 illustrates a flow chart for an embodiment of the method of the invention. -
FIG. 3 illustrates an example two-dimensional input image. -
FIG. 4 illustrates an example set of masks applied to areas or regions of the two-dimensional input image. -
FIG. 5A illustrates a perfect error free input depth map associated with the two-dimensional image.FIG. 5B illustrates a typical depth map having erroneous depth data as acquired from or calculated from an external depth capture apparatus or system. -
FIG. 6A illustrates a perfect normal map showing colors associated with angles of the surface of pixels within the two-dimensional image ofFIG. 5A .FIG. 6B illustrates an imperfect normal map associated with erroneous depth data as acquired from or calculated from an external depth capture apparatus or system as shown inFIG. 5B . -
FIGS. 7A-C illustrates a perspective, side and top view of perfect error-free depth applied to the two-dimensional input image.FIGS. 7D-F illustrates a perspective, side and top view of imperfect typical depth having errors and acquired from an external depth capture apparatus or system and applied to the two-dimensional input image. -
FIGS. 8A-C illustrates a perspective, side and top view of perfect error-free depth applied to planes and/or masks of areas or regions associated with the two-dimensional input image.FIGS. 8D-F illustrates a perspective, side and top view of imperfect typical depth acquired from an external depth capture apparatus or system and applied to planes and/or masks of areas or regions associated with to the two-dimensional input image. -
FIG. 9A illustrates a close-up view of the imperfect typical transformed into rotated planes which may or may not be flat planes, or curved planes or surfaces and which yield a manageable set of information that enables correction of the errors automatically by the system or through manual artist input accepted and processed by the system.FIG. 9B illustrates a close-up view of a portion ofFIG. 9B showing plane grouping with or without minimum and maximum connections and with or without spring attachments having a spring constant for example. -
FIG. 10 illustrates a flowchart of an embodiment of a method to convert 2D video to 3D video using an object model that defines the three-dimensional shape of an object that appears in one or more scenes of the 2D video. -
FIG. 11 illustrates an embodiment of the method shown inFIG. 10 , where the 3D object model is obtained from 3D scanner data captured from the object. -
FIG. 12 illustrates an example of steps fromFIG. 11 applied to a particular object in a video frame; 3D scanner data is converted to a polygon model, which is located in a frame, and the model is converted to a depth map. -
FIG. 13 illustrates an embodiment of a method for determining the position and orientation of a 3D object model in a frame by aligning features of the 3D model with corresponding pixels in the frame. -
FIG. 14 illustrates an embodiment of a method for tracking features across frames to propagate 3D model position and orientation from a key frame to other frames. -
FIG. 15 illustrates an embodiment of a 3D object model that has been rigged with a degree of freedom for a moveable joint. -
FIG. 16 illustrates tracking the 3D object model ofFIG. 15 across frames. -
FIG. 17 illustrates an extension to the example shown inFIG. 12 , with degrees of freedom added to the 3D object model. -
FIG. 18 illustrates an example of converting 3D scanner data to a planar model using planes fit to the points and links between the planes for constraints on plane boundaries. -
FIG. 19 shows an architectural diagram of a centralized machine learning system that performs multiple steps of a 2D to 3D conversion process, either fully automatically or semi-automatically with user input and review. -
FIG. 20 illustrates selected 2D to 3D conversion steps fromFIG. 19 , some or all of which may be performed by a machine learning system in one or more embodiments. -
FIG. 21 shows an example of 2D to 3D conversion using a machine learning system. -
FIG. 22 illustrates an architecture of a machine learning system that develops 2D to 3D conversion methods using a training set of conversion examples. -
FIG. 23 illustrates feature vector extraction for a machine learning method to perform object masking. -
FIG. 24 illustrates a machine learning method to learn a mapping between the feature vector ofFIG. 23 and an object mask defined by Bézier curves. -
FIG. 25 illustrates a machine learning method to learn a mapping between pixel values within an object mask and depth assignments to those pixels; example depth values may be obtained for example from a 3D point cloud or from a geometric model. -
FIG. 26 illustrates a machine learning method to learn a mapping between a series of 2D frames and a clean plate frame containing the background for those frames. - A method for converting 2D video to 3D video using machine learning will now be described. In the following exemplary description numerous specific details are set forth in order to provide a more thorough understanding of embodiments of the invention. It will be apparent, however, to an artisan of ordinary skill that embodiments of the invention may be practiced without incorporating all aspects of the specific details described herein. In other instances, specific features, quantities, or measurements well known to those of ordinary skill in the art have not been described in detail so as not to obscure the invention. Readers should note that although examples of the invention are set forth herein, the claims, and the full scope of any equivalents, are what define the metes and bounds of the invention.
-
FIG. 1 illustrates an exemplary overall system architecture for one or more embodiments of the invention. As shown,camera 101 and associated externaldepth capture apparatus 102, e.g., a single apparatus such as a light-field or time-of-flight apparatus, or as shown with two side mounted witness cameras for example, are utilized to capture images and depth values associated withforeground object 150,background object 151 inscene 152. Generally, multiple images are captured at fixed points in time to obtain a movie. Other embodiments may utilize depth captured from or input via other systems not directly coupled withcamera 101, including computer generated depths associated with computer-animated characters or objects, external LIDAR, plenoptic camera or any other external depth capture apparatus for example. Embodiments of the system generally include a computer such asserver 110 that enables artist(s) 160 to apply and/or correct depth for example as obtained from the externaldepth capture apparatus 102 or any other external source of depth. Embodiments of the server are generally utilized to obtain masks associated with areas within theforeground object 150,background object 151 or elsewhere in thescene 152. The masks may be automatically generated or accepted by the server for example byartist 160. The server may automatically calculate a fit for planes or masks associated with areas of the images or accept inputs to alter the fit fromartist 160 for example. The system may optionally also automatically alter the position, orientation, shape, depth or curve of planes or masks to fit the edges of the planes or masks with other planes or masks for example. Embodiments apply the depth of the planes or masks to the areas in the image to produce a stereoscopic image, e.g., anaglyph, or stereoscopic image pair for display onvisual output 120 byviewer 170. -
FIG. 2 illustrates a flow chart for an embodiment of the method of the invention wherein the method includes obtaining an external depth map associated with a two-dimensional image at 201, obtaining at least one mask associated with at least one area within the two-dimensional image at 202, calculating a fit or best fit for a plane using a computer based on depth associated with the at least one area associated with each of the at least one mask at 203, optionally, embodiments of the method may also automatically alter the position, orientation, shape, depth or curve of planes or masks to fit the edges of the planes or masks with other planes or masks for example at 204, applying depth associated with the plane having the fit to the at least one area to shift pixels in the two-dimensional image horizontally to produce a stereoscopic image or stereoscopic image pair at 205. One or more embodiments thus enable an external depth map transformation method of converting two-dimensional images to stereoscopic images that provides increased artistic and technical flexibility and rapid conversion of movies for stereoscopic viewing. Embodiments of the invention convert a large set of highly granular depth information inherent in a depth map associated with a two-dimensional image to a smaller set of rotated planes associated with masked areas in the image. This enables the planes to be manipulated independently or as part of a group and eliminates many problems associated with importing external depth maps including minimization of errors that frequently exist in external depth maps. -
FIG. 3 illustrates an example two-dimensional input image. As shown,foreground object 150, here a CGI human torso andbackground object 151, here a box form a scene in two-dimensional space with no depth information encoded into the image. -
FIG. 4 illustrates an example set of masks applied to areas or regions of the two-dimensional input image. As shown, masks 401 and 402 are utilized to represent sides of thebox 151 a with flat planes, whilemasks box 151 b is also shown with masks that are covered by foreground object, e.g., the human torso. Embodiments of the invention may also include obtaining at least one mask associated with at least one area within the two-dimensional image by automatically generating the at least one mask comprising the at least one area wherein the at least one area is over a predefined size and within a predefined depth range, or automatically generating the at least one mask comprising the at least one area wherein the at least one area comprises a boundary having a difference in luminance values over a predefined threshold, or both methods of size, depth range and boundary or any combination thereof. As shown masks representingflat planes -
FIG. 5A illustrates a perfect error free input depth map associated with the two-dimensional image. As shown, darker luminance values are encoded as nearer to the observer, although the reverse encoding may be utilized. There is no requirement that the depth map has to be in a viewable format, however, this encoding may enable intuitive viewing of depth errors for example. If all depth maps were error free, then embodiments of the invention may still be utilized to alter depth associated with the two-dimensional image in a manageable manner. However, since most depth maps are not perfect, embodiments of the invention enable error minimization and manageability unknown in the art. -
FIG. 5B illustrates a typical depth map havingerroneous depth data 501 as acquired from or calculated from an external depth capture apparatus or system. This problem makes some external depth maps nearly unusable for automated depth conversion, however, embodiments of the system enable use of error prone depth maps in a manageable manner. - Embodiments of the method may also include obtaining of the external depth map associated with a two-dimensional image by obtaining a disparity map, or a depth map of lower resolution than the two-dimensional image from a pair of witness cameras, or a depth map from time-of-flight system, or a depth map from a triangulation system. Embodiments of the invention may also include obtaining any type of depth map at 201 including Z-Depth associated with images that are generated through rendering from a Computer Generated Imagery or CGI application such as MAYA® or HOUDINI® as shown for example in
FIG. 5A , depth maps obtained after conversion of a disparity map from a stereoscopic pair of images to a Z-Depth for example as shown inFIG. 5B , Z-Depth extraction from of a light-field image, time-of-flight imaging systems, LIDAR, or any other type of depth map associated with a two-dimensional image. - Embodiments of the invention include a number of inherent advantages over simply using the Z-Depths as is currently performed in automated or semi-automated 2D to 3D conversion processes.
- For example, embodiments of the invention transform the large set of depth map depths or Z-Depth into a manageable number of parts. Thus, the system enables
artist 160 to manipulate individual or groups of parts for artistic purposes, as opposed to pixel-by-pixel editing. So, for example, the artist may independently adjust the angle, and hence depth of a robot's arm so the resulting stereoscopic image appears to reach out of the screen. In one or more embodiments, the planes may be grouped and movement or reshaping of a plane in two or three dimensions may move or reshape other grouped or otherwise coupled planes. - In addition, by transforming the Z-Depth into a manageable number of parts, the system enables an artist to group these parts and apply separate RGB image layers to these groups. This enables more efficient occlusion filling in the 2D to 3D conversion workflow.
-
FIG. 6A illustrates a perfect normal map showing colors associated with angles of the surface of pixels within the two-dimensional image ofFIG. 5A . For example by encoding separate RGB channels with X, Y and Z information related to the vector orthogonal to the particular area, a viewable format for normal vectors is thus obtained. If X increases from left to right, and Red is utilized as the X vector channel, then areas pointing to the right for example have a higher red value. If Y increases in the vertical direction and if Green is utilized for the Y channel, the areas having an orthogonal vector pointing have a higher green value. Any other encoding for visually viewable normal vector formats is in keeping with the spirit of the invention. -
FIG. 6B illustrates an imperfect normal map associated with erroneous depth data as acquired from or calculated from an external depth capture apparatus or system as shown inFIG. 5B . This image shows the severe problems with error prone depth data showingerror area 501 indicating an area that is pointing upward, when it is really not (seeFIG. 6A for error free portion of image). -
FIGS. 7A-C illustrate a perspective, side and top view of perfect error-free depth applied to the two-dimensional input image. If depth is error free, then the foreground object may be viewed with depth without errors in the foreground object, however, depending on the depth capture system, missing background data may exist and cause artifacts if there are not other images in a scene that have the missing background information. This occurs for example if the foreground object does not move with respect to the background during the scene. -
FIGS. 7D-F illustrate a perspective, side and top view of imperfect typical depth having errors and acquired from an external depth capture apparatus or system and applied to the two-dimensional input image. As shown, portions of the torso stretch outward when they should not, errors in theleft box 151 are shown as well with a curved flat surface near the top. The errors in the depth shown as applied inFIGS. 7D-F are difficult to fix by an artist since the errors occur on a small granular level, for example pixel or sub-pixel and require tedious paint style operations to correct. In addition, altering a depth of a visible area such as the arm is also quite difficult and economically unfeasible since the operations occur at the pixel level and not associated with a mask or masks that represents the arm for example. -
FIGS. 8A-C illustrate a perspective, side and top view of perfect error-free depth applied to planes and/or masks of areas or regions associated with the two-dimensional input image. If error free depth maps are obtained by the system, then applying the masks ofFIG. 4 results after calculation of plane rotations for example results in a readily editable image. -
FIGS. 8D-F illustrate a perspective, side and top view of imperfect typical depth acquired from an external depth capture apparatus or system and applied to planes and/or masks of areas or regions associated with to the two-dimensional input image. As shown, many of the errors of shown inFIGS. 7D-F are entirely or nearly entirely eliminated which enables rapid adjustment of depth of correction of depth errors if at all depending on the quality requirements of the particular project. For example, through calculation of best fit planes (whether flat or curved) at 203 as detailed above with respect toFIG. 2 , many of the errors are minimized or otherwise averaged away. - Whether using a perfect depth map that is error free or not, embodiments of the invention may also include calculating the best fit for a plane using a computer based on depth associated with the at least one area associated with each of the at least one mask by calculating a normal vector for the plane, or a regression fit for the plane, or an orthogonal distance regression fit for the plane, or in any other known manner regarding fitting a plane to particulars points in three-dimensional space. Specifically, embodiments of the invention mold depth data to eliminate depth errors by transforming large numbers of depth values to smaller number of plane rotations. In one embodiment, the system may calculate the normal and position for a specific region, for example to form an average, rotation value associated with a plane that represents a large group of depth values, some of which may be erroneous. Hence, issues associated with imperfect depth map data are often averaged out, or otherwise eliminated. In some extreme cases of noisy depth data, these issues may not be fully resolved, however, embodiments of the invention reduce the problem to a manageable number of editable parts, and enable the issues to be rapidly and easily corrected automatically or by accepting inputs from an artist. Although embodiments of the invention may utilize a normal vector approach, other algorithms may be utilized alone or in combination to achieve similar or advantageous results. For example, embodiments of the invention may treat each pixel as a point in space, e.g., wherein X and Y represent the position of the pixel and Z represents the Z-Depth value of that pixel, and isolate only the points within the defined region, and calculate the “best-fit” plane for that group of points. Embodiments of the invention may utilize any type of plane fitting algorithm including but not limited to regression plane, orthogonal distance regression plane, etc. Commonly available statistics toolboxes include orthogonal regression using principal components analysis for example that may be utilized as off the shelf software components for calculation of best fitting planes to a number of points for example to minimize the perpendicular distances from each of the points to a plane. Embodiments of the invention may utilize any type of filtering as part of the transformation processing including but not limited to dilation and erosion. In one or more embodiments, an algorithm that iterates over a set of depth slopes and averages the slopes over an area for example is one example of an algorithm that may be utilized to calculate the normal vector for a particular area of the depth map.
-
FIG. 9A illustrates a close-up view of the imperfect typical transformed into rotated planes which may or may not be flat planes, or curved planes or surfaces and which yield a manageable set of information that enables correction of the errors automatically by the system or through manual artist input accepted and processed by the system. As shown, even with a high degree of erroneous depth data, the masks are low in number, and thus are far easier to adjust to correct depth associated with the two-dimensional image. -
FIG. 9B illustrates a close-up view of a portion ofFIG. 9B showing plane grouping with or without minimum and maximum connections and with or without spring attachments having a spring constant for example. In one or more embodiments, plane grouping may be visibly shown and editable to enable grouped planes to be repositioned by accepting a drag or other user input wherein thecouplings - Alternatively or in combination, embodiments of the invention may include altering automatically any combination of position, orientation, shape, depth or curve of said plane in order to fit edges or corners of the plane with another plane. This enables a plane to be positioned in three-dimensional space, rotated in three-dimensions, reshaped by moving a corner point, warped in effect by adding depth or a curve to the plane, for example to add depth to the plane itself to match the underlying image data. Embodiments of the invention may also include accepting an input to alter any combination of position, orientation, shape, depth or curve of the plane, for example to artistically fit the underlying image data, correct errors or artifacts from the automated fitting process for touch up, etc.
- Embodiments of the invention generally also include applying depth associated with the plane having the best fit to the at least one area to shift pixels in the two-dimensional image horizontally to produce a stereoscopic image or stereoscopic image pair. Any type of output that is capable of providing different left and right eye information is in keeping with the spirit of the invention.
- One or more embodiments of the invention may use external depth information incorporated into a 3D model of one more objects that appear in a scene. A 3D object model may for example provide depth for the portions of an object, rather than for an entire image or an entire sequence of frames in a video. By locating and orienting the 3D object model in each image or frame, one or more embodiments may simplify the task of generating a depth map for an entire image or for an entire video.
-
FIG. 10 illustrates a flowchart of an embodiment of a method that uses a 3D object model to convert a 2D video to a 3D video. A2D video 1001 is obtained instep 1000. Thevideo 1001 may contain one or more scenes, where a scene is for example associated with a particular location for a shot. In step 1010 a3D space 1011 is defined for one of the scenes of thevideo 1001. This 3D space may for example define a coordinate system for the location of the scene. To convert a2D video 1001 to 3D, objects in the scene must be located in this3D space 1011. Techniques known in the art for 2D to 3D conversion often require manual depth assignment to every pixel or to every region of each frame. This process is time-consuming and labor-intensive. In some situations, significant efficiencies may result from modeling the 3D shape of one or more objects once, and applying these 3D object models across frames. In the embodiment ofFIG. 10 , a3D object model 1021 is obtained. One or more embodiments may obtain any number of 3D object models. 3D object models may be obtained from any source and in any format. For example, without limitation, 3D object models may comprise point clouds, polygonal models, depth maps, disparity maps, parallax maps, parameterized surfaces such as NURBS surfaces, or 3D models in any of the file formats used in the art such as for example 3DS, FBX, OBJ, STEP, STL, DAE, X3D, BLEND, or others. In one ormore embodiments 3D object models may be combinations of any of the previous examples or of any other 3D model formats. - In
step 1030 the frames containing the object are identified. For example, inFIG. 10 , frames 1031, 1032 and 1033 ofvideo 1001 contain images of the object with3D object model 1021. In one or more embodiments frames containing the object or objects of interest may be determined manually. In one or more embodiments frames containing the object or objects of interest may be determined automatically using a computer. For each of these frames the embodiment shown inFIG. 10 uses the3D object model 1021 to convert the frame to 3D. Instep 1040, the object position and orientation in the frame is determined. This results in placing theobject model 1021 in3D space 1011, resulting in transformedobject model 1041. From this transformed3D object model 1041,step 1050 generates adepth map 1051 for the object in the frame. Depth map generation from the transformed 3D object model is a straightforward calculation since the location of each point of the object is known in the3D space 1011. Instep 1060,depth 1061 is assigned to additional elements, if any, in the frame. In one or more embodiments multiple objects of interest may exist in a frame, and multiple 3D object models may be applied to the frame to generate individual object depth maps; these individual object depth maps may then be combined into an overall depth map. In one or more embodiments additional 3D modeling or depth editing steps may be used to complete the depth map for a frame. In one or more embodiments the 3D object model may comprise the entire frame, so that additional depth map generation is not required. Instep 1070, a stereoscopic image pair is generated from the frame; this image pair comprises leftimage 1071 andright image 1072. Using techniques known in the art, pixels may be shifted horizontally between the left and right images based on thedepth map 1061. One or more embodiments may use any technique known in the art to generate stereoscopic image pairs from depth maps or from other 3D model information. For example, one or more embodiments may use any of the techniques known in the art for filling of gaps created by pixel shifting. - One or more embodiments may obtain one or more 3D object models from any source or sources and in any format or formats. One potential source for 3D object models is 3D scanner data. 3D scanner data may for example be obtained from an object and converted into a 3D object model used for 2D to 3D conversion. Any object may be used to generate a 3D object model, such as for example, without limitation, a person, a group of persons, an animal, an inanimate object, a machine, a building, a vehicle, a computer-generated figure, or a physical model of any other object.
FIG. 11 illustrates an embodiment of this method.Step 1020, obtain 3D object model, comprises two steps:step 1025 obtains3D scanner data 1101 of an object, and step 1026 converts this 3D scanner data to alower polygon count 3D object model 1102. 3D scanner data may be obtained using any 3D scanning technology. Technologies known in the art include for example contact and non-contact methods. Non-contact methods include for example laser time-of-flight methods, laser triangulation methods, holographic methods, and structured light methods. Any or all of these methods, or any other 3D scanning methods, may be used as part of the process of obtaining a 3D model of an object. Embodiments may obtain 3D scanner data in any format, including for example, without limitation, point clouds, polygonal models, NURBS surface models, or 3D models in any of the formats known in the art. One or more embodiments may use 3D scanner data directly as a 3D object model. One or more embodiments may convert 3D scanner data to a different format. For example, a point cloud may be converted to a polygon model or a parameterized surface model using techniques known in the art. One or more embodiments may convert 3D scanner data to alower polygon count 3D model for efficiency of processing, as illustrated instep 1026. Reducing the polygon count of a 3D model is known in the art as retopologizing; one or more embodiments may use any retopologizing methods or technologies known in the art to convert 3D scanner data or other 3D models to alower polygon count 3D model. One or more embodiments may use 3D models that comprise parameterized surfaces, such as NURBS surfaces for example, instead of or in addition to polygons. -
FIG. 12 illustrates an example of selected steps of the embodiments shown inFIGS. 10 and 11 . A3D scanner model 1101 a is obtained of an object, as instep 1025 ofFIG. 11 . This 3D scanner model may for example comprise a very large number of polygons, which may for example have been generated from a point cloud from a 3D scanner. This 3D scanner model is converted to alower 1102 a, shown as an overlay on the 3D scanner model, as inpolygon 3D modelstep 1026 ofFIG. 11 . Themodel 1102 a is located and oriented at position andorientation 1041 a inframe 1031 a; this step overlays and aligns the 3D object model onto the image of the character in the frame. From thismodel 1041 a situated and oriented in the 3D space associated with the scene,depth map 1051 a is generated for the character. Here depth has not yet been assigned to other elements of the scene in the frame, as indicated by the white background surrounding the character. This process of aligning the3D model 1102 a with the image of the character in the frame may be repeated for every frame in the scene containing the character. This use of the 3D model therefore greatly facilitates depth map generation for the scene. - One or more embodiments may use any technique or techniques to determine the position and orientation of a 3D object model in a frame. These techniques may be manual, automated, or a combination of manual and automated techniques.
FIG. 13 illustrates an embodiment of a method that aligns a 3D object model with an image using one or more features. Features may be for example points or regions within an object model that have distinguishing shapes, corners, lines, textures, colors, or other characteristics. In the example shown inFIG. 13 , fourfeatures frame 1031 a at pixel coordinates 1311, 1312, 1313, and 1314, respectively. The correspondence between feature positions in the 3D object model and the corresponding pixel coordinates in the frame allows the position and orientation of the object model in the 3D space to be calculated. Techniques for recovering an object pose from the pixel coordinates of its features are known in the art. In general, a unique object pose may be determined from the pixel coordinates of 4 features. For arbitrary numbers of feature points these techniques are known in the art as solutions to the “Perspective-n-Point problem,” since they recover the object pose from the projection of n points (features) of the object onto an image plane. For more than 4 features, object pose may be overdetermined, in which case many techniques perform an error minimization to best align an object pose with the projection of the features onto an image. See for example: D. DeMenthon and L. Davis (1995). Model Based Object Pose in 25 Lines of Code. International Journal of Computer Vision, 15(1-2), 123-141. One or more embodiments may use any of the techniques known in the art to obtain or calculate the position and orientation of a 3D object model from the pixel location of any number of model features in an image projection in a frame. - One or more embodiments may locate 3D object model features in one or more frames manually, automatically, or using a combination of manual and automated methods.
FIG. 14 illustrates an embodiment that tracks a feature across frames using manual alignment in a key frame with automated tracking of a feature in non-key frames. In this example a3D object model 1403 of the head of a grizzly bear is obtained, and a feature is selected that corresponds to the tip of the bear's nose. Inkey frame 1401 the feature is manually located atpixel location 1404 for example by an operator. Innon-key frame 1402 thepixel location 1405 of the feature is located automatically by searching pixels for characteristics corresponding to the feature. Techniques for automatic feature tracking are known in the art; one or more embodiments may use any known techniques for tracking features of a 3D object model across frames.Feature track 1406 indicates how the feature pixel location moves across frames. The orientation and position of the3D object model 1403 in each frame may be determined by tracking multiple features, as described above. -
FIG. 14 illustrates a potential issue with a rigid 3D object model. While theobject model 1403 matches the object (the bear's head) fairly well in thekey frame 1401, inframe 1402 the bear'sopen jaw 1407 causes the model to deviate from the image. One or more embodiments may use 3D object models with degrees of freedom, for example for moveable joints, to match object images for non-rigid objects.FIG. 15 illustrates a simplified object model for thebear 1500 that incorporates a degree of freedom for the jaw. (Here the model is shown in 2 dimensions for ease of illustration.) The object model is divided into twoobject model parts freedom 1503 between the two parts, which represents the angle of the jaw. In one or more embodiments, a 3D object model may be modified to incorporate multiple parts and degrees of freedom for relative motion of the parts. This process is known in the art as rigging the 3D object model. One or more embodiments may incorporate one or more rigging steps into the method to account for potential degrees of freedom in the object. - With a rigged 3D object model, aligning the object model with an image in a frame involves positioning and orienting each part of the model. In general, this alignment determines an overall position and orientation for the model, along with values for each degree of freedom. One or more embodiments may use manual, automated, or mixed methods to determine the model orientation and position and the values of each degrees of freedom in any frame. One or more embodiments may manually set the values for each degree of freedom in one or more key frames, and use automated or semi-automated methods to determine these values in non-key frames. For example, one or more embodiments may automatically interpolate values for degrees of freedom between key frames. One or more embodiments may use features on each or a subset of the model parts, and use automated feature tracking for the parts to determine values for the degrees of freedom.
FIG. 16 illustrates this process for the example frames shown inFIG. 14 , using the rigged 3D object model fromFIG. 15 . Inkey frame 1401 an operator aligns the 3D object model with the image and sets the rotational degree offreedom 1602 to zero, representing the closed jaw of the bear. A feature is selected for each part of the model;feature 1404 is the tip of the nose, as before, and feature 1601 is the front tip of the lower jaw for the bottom part of the model. In the key frame these features are adjacent. Automated feature tracking then tracks these features to thenon-key frame 1402;feature 1404 moves tolocation 1405, and feature 1601 moves tolocation 1603. From these new feature positions, the rotation degree offreedom 1604 may be automatically calculated. -
FIG. 17 illustrates a rigged 3D object model applied to the example fromFIG. 12 . The static3D object model 1102 a is rigged to generate a rigged3D object model 1701, with degrees of freedom added for the possible motion of the character's joints. For example, degree offreedom 1702 represents potential tilting of the neck. Inframe 1041 a, thevalue 1703 for this neck tilt is calculated. This value may be set manually by an operator to align the model parts with the image or as discussed above it may be determined automatically using for example feature tracking of features for the individual model parts. - In converting 3D scan data to a 3D object model, one or more embodiments may fit one or more planes to the scan data using any or all of the techniques described above.
FIG. 18 illustrates an example of an embodiment that generates planes to fit3D scan data 1101 a. For example,plane 1801 is fit to the cheek,plane 1802 is fit to the jaw, and alink 1803 is defined between these two planes. The link may for example define minimum and maximum distances or angles, or a spring constant. Planes may be fit automatically to the data using constraints defined by the links. After the initial fit, the edges, angles, or normal of the planes may for example be adjusted to fit the edges or corners of the planes with one another to eliminate gaps or misalignments in the resulting 3D object model. - One or more embodiments may use machine learning to execute any or all conversion steps in converting a 2D video to a 3D video. In the art, these steps for 2D to 3D conversion are often performed manually, sometimes with assistance from specialized tools. One or more embodiments of the invention enable a method for performing one or more of these steps using a machine learning system. This system may dramatically reduce the time and cost of 2D to 3D conversion. A machine learning system may also be used to track complexity, costs, budgets, status, capacity, and workload, and to forecast these variables for projects or tasks.
FIG. 19 shows an architectural overview of an embodiment of the system that uses a machine learning system to perform and coordinate various tasks associated with 2D to 3D conversion. The centralizedmachine learning system 1900 coordinates the flow of information throughout the process. It may access data from previous conversion projects to develop and refine machine learning algorithms to perform the conversion steps. The steps may be performed fully automatically bymachine learning system 1900, or operators may assist with review and revisions.2D video frames 1901 are provided to themachine learning system 1900; the system then executes a series of steps to convert from 2D to 3D. Outputs of each step are fed back into the machine learning system for possible checking, correction, refinement, and adaptive tuning of the conversion algorithms. The steps shown inFIG. 19 are illustrative; one or more embodiments may use a machine learning system to perform additional conversion steps, or to perform steps in a different order.Image Recognition process 1910 identifies objects and components within the 2D frames 1901. This process may for example use pattern recognition, known object templates, known object features, or classifiers to determine the objects that are present in a frame. The Image Recognition process may also determine the location and the edges for each detected object. Thesubsequent Tagging process 1920 assigns identities to the subjects recognized by theImage Recognition process 1910. For example, the Tagging process may identify specific actors, or objects associated with broad categories of objects such as people, animals, vehicles, or buildings. The Tagging process may also assign one or more attributes to each identified object.Grouping step 1930 may combine similar frames or similar shots, for example based on the frames or shots having similar objects within the frames. 2D-3D Conversion Step 1940 generates 3D stereo image pairs from the 2D Frames 1901. For example, this step may assign or create masks for the objects identified inImage Recognition step 1910 and tagged inTagging step 1920. It may also apply depth to the masked areas, using known depth models for specific types of objects or using machine learning algorithms developed from previous conversions. Theresults 1970 include the stereo image pair outputs, as well as statistics for the conversion process such as complexity of the masks used and the time needed to perform the conversions. All outputs from steps such as 1910, 1920, 1930, 1940, and 1970 may be fed back intoMachine Learning System 1900 for revision, correction, validation, or refinement. These feedback processes also allow the system to develop adaptive algorithms that learn from previous conversion examples. - Generation of stereo images by 2D-
3D Conversion process 1940 may in some cases create gaps in images as objects are shifted left and right based on depth. TheMachine Learning System 1900 may performGap Filling 1960 to fill these gaps. For example, the system may generate Clean Plate images that contain background elements only, and extract pixels from the Clean Plate to fill the gaps. Clean Plates may be generated for example usingObject Tracking 1950, which allows the system to identify moving objects and to separate stationary objects from moving objects in each frame. - Centralized
Machine Learning System 1900 may also perform project planning, budgeting, tracking, and forecasting. For example, users may input preliminary estimates for budgets andproject complexity 1980. As the conversion process proceeds, the system may generate and updateforecasts 1990 for project variables, such as for example required capacity, costs, and bid times. -
FIG. 20 shows illustrative 2D to 3D conversion steps that may for example be performed fully or partially by a Machine Learning System such as the system shown inFIG. 19 . These steps are illustrative; one or more embodiments may use any subset or any superset of these steps to convert any 2D image or 2D video into 3D.2D video 2001 is an input into the conversion process. This video comprises one or more 2D frames, such asframe 2002.Step 2010 identifies and locates one or more objects in one or more of these frames. For example,object 2011 is located and identified inframe 2002. The followingstep 2020 masks the object (or objects). For example, rotoscoping is a technique known in the art for masking objects. The resultingobject mask 2021 identifies the pixels in theframe 2002 that comprise the image of theobject 2011. Binary masks are typically used in the art to represent object masks; one or more embodiments may use any type of mask. The followingstep 2030 models the depth of the masked object or objects. This step creates a 3D model or a depth map that indicates how far away each pixel within the mask is from the viewer. In the illustrative example ofFIG. 20 , theobject 2011 is modeled as asphere 2031 within3D space 2032. One or more embodiments may use any geometric model or models to represent the depth or 3D shape of an object. Once depth has been assigned to the objects in a frame,step 2040 generates astereoscopic image pair step 2040. Typically, pixels are shifted horizontally between left and right images, with the amount of shifting inversely proportional to the depth of the pixel from the viewer. Often this process will reveal portions of an image that were occluded by the objects being shifted, resulting in missing pixels in the left or right images. For example, inleft image 2041, theobject 2011 has been shifted to the right, leaving agap 2043 with missing pixels.Step 2050 fills the gaps (if any) created by pixel shifting. For example,image 2051 haspixels 2052 inserted into thegap area 2043. Techniques for gap filling are known in the art; one or more embodiments may use any of these techniques to create values for missing pixels. The complete stereo image pair with gaps filled then becomes a 3D frame in3D video 2060, which is an output of the conversion process. -
FIG. 21 shows an example of the steps ofFIG. 20 applied to a frame. These steps may be performed fully or partially by a machine learning system, using for example the techniques described in the following sections.Frame 2002 a is a 2D frame.Steps frame 2002 a. For example, the head of the actor is identified withtags 2011 a and is masked withmask boundary 2021 a; the body of the actor is identified withtags 2011 b and is masked withmask boundary 2021 b; and the background is identified withtags 2011 c and is masked withmask boundary 2021 c. TheModel Depth step 2030 then assigns depth to the pixels in each of the masked regions corresponding to the identified objects. In this illustrative example depths are represented as grayscale values with darker values being closer to the viewer. The head object hasdepth values 2031 a; the body object hasdepth values 2031 b; and the background object hasdepth values 2031 c. Note that the depth values of the background are not uniform; since the Id & LocateObject process 2010 tagged the background as comprising “leaves” and “bushes” the system may apply a non-uniform depth to this object to reflect the natural depth variations associated with this type of object. The GenerateStereo step 2040 then creates astereoscopic image pair depth values -
FIG. 22 shows an embodiment of a machine learning system that develops and improves 2D to 3D conversion algorithms by training the system on conversion examples. Atraining set 2201 of 2D to 3D conversion examples is used to trainmachine learning system 1900. The training set comprises a (potentially very large) set of conversion examples such as conversion example 2202. A conversion example may include a2D input 2203, a3D output 2204, and data representing one or more conversion steps 2205. For example, conversion examples may include any or all of the conversion steps described inFIG. 19 orFIG. 20 , or illustrated in the example ofFIG. 21 . Conversion examples may also include any other conversion steps, preprocessing steps, or post-processing steps. One or more embodiments may use any format for describing the inputs, transformations, and outputs of any conversion step. Thetraining process 2206 may be iterative, with themachine learning system 1900 updated and refined at each iteration to improve its performance on the training set. As is known in the art, one or more embodiments may also use a test set and a validation set to assist with training, tuning, and selection of themachine learning system 1900. - Machine learning systems and methods for training these systems are known in the art. One or more embodiments may use any of these machine learning techniques to design and train
machine learning system 1900. For example, without limitation,machine learning system 1900 may include neural networks, deep learning systems, support vector machines, regression models, classifiers, decision trees, ensembles, genetic algorithms, hidden Markov models, and probabilistic models. One or more embodiments may use supervised learning, unsupervised learning, semi-supervised learning, deep learning, or combinations thereof. One or more embodiments may preprocess training data in any desired manner, for example using feature vector selection, dimensionality reduction, or clustering. One or more embodiments may use any optimization technique or techniques to select and refine parameters for the machine learning system. - After
machine learning system 1900 is trained, it may be applied to convert a2D video 2220 to a3D video 2221. The conversion steps may include, for example, without limitation, any or all of thesteps FIG. 20 . The steps may also include, without limitation, any or all of thesteps FIG. 19 . One or more embodiments may applymachine learning system 1900 to any or all of these conversion steps, and to any additional conversion steps. The machine learning system may perform any step completely, or partially. In one or more embodiments anoperator 2230 may manually perform selected steps, or the operator may modify, check, revise, or complete selected steps that have been performed bymachine learning system 1900. In one or more embodiments themachine learning system 1900 and theoperator 2230 may interact to perform, review, or modify any or all of the 2D to 3D conversion steps. The steps may be performed in any order, and may be iterative. One or more steps may be optional. One or more steps may be performed multiple times; for example, object masking may be performed for multiple objects within a scene. - We now describe embodiments of a machine learning system applied to the steps 2020 (Mask Object), 2030 (Model Depth), and 2050 (Fill Gaps). These embodiments are illustrative; one or more embodiments may use any machine learning techniques applied to any or all of the steps required to convert a 2D video to a 3D video. Machine learning techniques to learn any desired function from an input space to an output space are known in the art. The discussion below therefore describes illustrative representations for the inputs and outputs of the 2D to 3D conversion steps. Well-known machine learning techniques may be used to train a machine learning system using a set of training examples comprising these inputs and outputs.
-
FIG. 23 illustrates an embodiment that uses machine learning to mask an object. An initial object identification andlocation step 2010 is first performed on2D frame 2301; this step may be performed manually, using a machine learning method, using image processing and pattern recognition techniques known in the art, or using any combination of these methods. In this example, the frame has been analyzed and anidentity 2302 of an object in the frame has been calculated or obtained. One or more embodiments may use any desired granularity for object identities. For example, objectidentity 2302 could be more general, such as “human,” or more specific, such as “Alice.” Based on the object identity, a set of object features 2303 is associated with the object. These features may for example correspond to specific identifiable locations within an image of the object.Step 2010 selects these features based on theobject identity 2302, and locates the pixel coordinates of each feature inframe 2301. Different features may be selected for different object identities. For example, an object type of “cat” might have an associated “tail” feature, while an object type of “truck” might have an associated “left front wheel” feature. - In the illustrative embodiment of
FIG. 23 , the feature points 2304 a, 2304 b, and 2304 c are used to create a larger grid ofpoints 2305, which is then used to generate an object mask. This approach is illustrative; one or more embodiments may use any method to train a machine learning system to generate an object mask for an identified object. In this example thepoints barycentric coordinates 2306. This approach generates a grid of points surrounding the feature points identified instep 2010. The pixel values for these grid points form afeature vector 2307. In this example the pixel values are expressed in HSV color space; one or more embodiments may use any desired color or grayscale space to represent pixel values. Thefeature vector 2307 serves as an illustrative input for the Mask Object function of the machine learning system. -
FIG. 24 continues the example ofFIG. 23 , and shows an illustrative output for theMask Object process 2020. In this example, an object mask is represented as asequence 2401 of connected Bézier curves that form a boundary for the masked region. This is an illustrative example; one or more embodiments may use any desired representation for a masked region, including for example using curves or segments of any shape and parameterization. The curves illustrated in 2401 are second order curves; each curve is defined by three control points. One or more embodiments may use curves of any order to represent a region boundary. For example,curve 2402 is defined by the threecontrol points control points 2404 comprises the output of theMask Object process 2020. With well-defined representations for both theinput Feature Vector 2307, and theOutput Vector 2404, the machine learning system can be trained to reproduce the Mask Object function 2020 from the sequence of training examples comprising input and output pairs for this step. -
FIG. 25 continues the example ofFIG. 24 to illustrate an approach to using machine learning for theModel Depth step 2030. This step generates a depth model for a masked object region. The illustrative input for this step shown inFIG. 25 isFeature Vector 2501, which includes pixel values at the grid points 2305 and at points on the boundary curves of the masked region as calculated inFIG. 24 . For example, the points on the boundary curves may comprise the endpoints of each curve together with intermediate points on each curve between the endpoints. The combinedgrid 2502 provides an illustrative sampling of the image across the masked region. TheFeature Vector 2501 comprises the pixel values at each of these grid points. As inFIG. 23 , pixel values are shown in HSV space, but one or more embodiments may use any desired pixel value representation. Theoutput vector 2503 is the depth (z value) of each grid point, based on a 3D depth model associated with the conversion example.FIG. 25 illustrates two potential sources for these depth values.Source 2504 is for example a 3D point cloud associated with the object. This point cloud may for example be obtained from a 3D scan of the object. Interpolation of depth values between points in the point cloud may be used to generate the grid depth values 2503.Source 2505 is for example a geometric surface model generated for the object. This model may for example comprise planar or curved regions that represent the shape of the object. It may be generated by a depth artist manually or using semi-automated tools. The depth of the points in thegrid 2503 may be calculated directly from the geometric model. The machine learning system may then estimate theModel Depth function 2030 from the set of training examples that compriseinput Feature Vectors 2501 andOutput Vectors 2503. In one or more embodiments a subsequent step may be performed to generate a depth model for all points in the masked region. For example, thegrid point depths 2503 could be used to generate a triangulated surface for the entire masked region. -
FIG. 26 illustrates an example of an embodiment that uses machine learning methods for theFill Gaps step 2050. This illustrative approach uses a clean plate, which is a frame that comprises only background objects, with all foreground objects removed. Methods for generating clean plates are known in the art, but these methods typically use manual procedures. In the example ofFIG. 26 , a machine learning system creates a clean plate automatically by identifying and tracking moving objects, and removing these from the background of multiple frames.Frames 2601 contain a stationary background and a foreground object that moves across frames.Process 2010 identifies the moving object and locates it in at least a key frame, as described earlier. This illustrative example shows a single moving object; one or more embodiments may use similar methods to identify multiple moving objects and to track them across frames.Process 1950 tracks the object location across the frames, for example using feature tracking techniques known in the art. For example,feature 2611 corresponds to the edge of the character's mouth in the first frame; theobject tracking process 1950 generatesfeature locations FIG. 24 , the pixels associated with the moving object in each frame can then be identified.Step 2620 then removes the pixels associated with the moving object from the frames, resulting inframes 2621 with just the background present. The GenerateClean Plate step 2630 then merges these background frames intoclean plate 2640. Using the clean plate, the machine learning system can then automatically fill gaps such asgaps 2043 inframe 2041. TheFill Gaps process 2050 takespixels 2641 from theclean plate 2640 to fill thearea 2043, resulting in afinal frame 2650 with no gaps. - As illustrated in
FIGS. 23 through 26 , one or more embodiments may use machine learning methods to learn and perform any or all of thesteps Mask Object 2020,Model Depth 2030, and FillGaps 2050. These examples are illustrative. Embodiments may use machine learning for any portion or portions of the 2D to 3D conversion process. In one or more embodiments machine learning techniques may be combined with operator supervision and input to form a combined machine-human conversion system. - In one or more embodiments machine learning methods may be combined with other automated or semi-automated processes to perform a 2D to 3D conversion. For example, a 3D object model may be available for one or more objects in a 2D scene. In this case it may not be necessary or desirable to use a machine learning system to perform the
Model Depth step 2030 for those objects, since the 3D object models already contain the depth values for the objects. The machine learning system may be used to model depth for other objects in a scene for which there is not 3D object model. It may also be used for other steps, such as for example theFill Gaps step 2050. - While the invention herein disclosed has been described by means of specific embodiments and applications thereof, numerous modifications and variations could be made thereto by those skilled in the art without departing from the scope of the invention set forth in the claims.
Claims (14)
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
US14/967,939 US9609307B1 (en) | 2015-09-17 | 2015-12-14 | Method of converting 2D video to 3D video using machine learning |
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
US14/857,704 US9438878B2 (en) | 2013-05-01 | 2015-09-17 | Method of converting 2D video to 3D video using 3D object models |
US14/967,939 US9609307B1 (en) | 2015-09-17 | 2015-12-14 | Method of converting 2D video to 3D video using machine learning |
Related Parent Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
US14/857,704 Continuation-In-Part US9438878B2 (en) | 2013-05-01 | 2015-09-17 | Method of converting 2D video to 3D video using 3D object models |
Publications (2)
Publication Number | Publication Date |
---|---|
US20170085863A1 true US20170085863A1 (en) | 2017-03-23 |
US9609307B1 US9609307B1 (en) | 2017-03-28 |
Family
ID=58283615
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
US14/967,939 Active US9609307B1 (en) | 2015-09-17 | 2015-12-14 | Method of converting 2D video to 3D video using machine learning |
Country Status (1)
Country | Link |
---|---|
US (1) | US9609307B1 (en) |
Cited By (34)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20170109888A1 (en) * | 2015-10-20 | 2017-04-20 | Hewlett-Packard Development Company, L.P. | Alignment of images of a three-dimensional object |
US10068385B2 (en) | 2015-12-15 | 2018-09-04 | Intel Corporation | Generation of synthetic 3-dimensional object images for recognition systems |
US20180295335A1 (en) * | 2017-04-10 | 2018-10-11 | Red Hen Systems Llc | Stereographic Imaging System Employing A Wide Field, Low Resolution Camera And A Narrow Field, High Resolution Camera |
US10366499B2 (en) * | 2016-09-23 | 2019-07-30 | Robotis Co., Ltd. | Device, method, and computer-readable recording medium for editing and playing robot motion |
WO2019209008A1 (en) * | 2018-04-24 | 2019-10-31 | 주식회사 지디에프랩 | System for improving video quality by using changed macroblock extraction technique |
KR20190130480A (en) * | 2018-04-24 | 2019-11-22 | 주식회사 지디에프랩 | Video resolution enhancement system using change macro-block extraction technique |
WO2020032354A1 (en) * | 2018-08-06 | 2020-02-13 | Samsung Electronics Co., Ltd. | Method, storage medium and apparatus for converting 2d picture set to 3d model |
US10657712B2 (en) * | 2018-05-25 | 2020-05-19 | Lowe's Companies, Inc. | System and techniques for automated mesh retopology |
US20200186782A1 (en) * | 2018-12-05 | 2020-06-11 | Fotonation Ltd. | Depth sensing camera system |
US10810738B1 (en) * | 2018-12-07 | 2020-10-20 | Bellus 3D, Inc. | Marker-less alignment of digital 3D face and jaw models |
CN111935528A (en) * | 2020-06-22 | 2020-11-13 | 北京百度网讯科技有限公司 | Video generation method and device |
US10839481B1 (en) * | 2018-12-07 | 2020-11-17 | Bellus 3D, Inc. | Automatic marker-less alignment of digital 3D face and jaw models |
WO2020231640A1 (en) * | 2019-05-13 | 2020-11-19 | Light Field Lab, Inc. | Light field display system for performance events |
CN112019828A (en) * | 2020-08-14 | 2020-12-01 | 上海网达软件股份有限公司 | Method for converting 2D (two-dimensional) video into 3D video |
US10991156B2 (en) * | 2018-12-05 | 2021-04-27 | Sri International | Multi-modal data fusion for enhanced 3D perception for platforms |
CN113177908A (en) * | 2021-04-01 | 2021-07-27 | 柳城县迪森人造板有限公司 | Identification method and device for solid wood ecological plate |
US20210235067A1 (en) * | 2018-07-06 | 2021-07-29 | Pcms Holdings, Inc. | Method and system for forming extended focal planes for large viewpoint changes |
US11100658B2 (en) * | 2020-01-10 | 2021-08-24 | United States Of America As Represented By The Secretary Of The Navy | Image georectification with mobile object removal |
US20210352261A1 (en) * | 2020-05-11 | 2021-11-11 | Niantic, Inc. | Generating stereo image date from moncular images |
US20220028162A1 (en) * | 2019-11-26 | 2022-01-27 | Sdc U.S. Smilepay Spv | Systems and methods for constructing a dental arch image using a machine learning model |
US20220051372A1 (en) * | 2020-08-12 | 2022-02-17 | Niantic, Inc. | Feature matching using features extracted from perspective corrected image |
CN114120149A (en) * | 2021-11-09 | 2022-03-01 | 肇庆市城市规划设计院 | Oblique photogrammetry building feature point extraction method and device, electronic equipment and medium |
US20220132222A1 (en) * | 2016-09-27 | 2022-04-28 | Clarifai, Inc. | Prediction model training via live stream concept association |
US20220189130A1 (en) * | 2017-11-29 | 2022-06-16 | Sdc U.S. Smilepay Spv | Systems and methods for constructing a three-dimensional model from two-dimensional images |
US11393164B2 (en) | 2019-05-06 | 2022-07-19 | Apple Inc. | Device, method, and graphical user interface for generating CGR objects |
WO2022178293A1 (en) * | 2021-02-19 | 2022-08-25 | Insurance Services Office, Inc. | Computer vision systems and methods for supplying missing point data in point clouds derived from stereoscopic image pairs |
US11470298B2 (en) * | 2019-11-11 | 2022-10-11 | Samsung Electronics Co., Ltd. | Method and apparatus with updating of algorithm for generating disparity image |
EP4070249A4 (en) * | 2020-12-16 | 2023-03-29 | Tencent America LLC | Reference of neural network model for adaptation of 2d video for streaming to heterogeneous client end-points |
US11689709B2 (en) | 2018-07-05 | 2023-06-27 | Interdigital Vc Holdings, Inc. | Method and system for near-eye focal plane overlays for 3D perception of content on 2D displays |
US11783539B2 (en) | 2019-05-17 | 2023-10-10 | SmileDirectClub LLC | Three-dimensional modeling toolkit |
US11850113B2 (en) | 2019-11-26 | 2023-12-26 | Sdc U.S. Smilepay Spv | Systems and methods for constructing a three-dimensional model from two-dimensional images |
US11871137B2 (en) * | 2020-09-30 | 2024-01-09 | Lemon Inc. | Method and apparatus for converting picture into video, and device and storage medium |
US11880990B2 (en) | 2020-09-21 | 2024-01-23 | Samsung Electronics Co., Ltd. | Method and apparatus with feature embedding |
US11893755B2 (en) | 2018-01-19 | 2024-02-06 | Interdigital Vc Holdings, Inc. | Multi-focal planes with varying positions |
Families Citing this family (13)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US9324190B2 (en) | 2012-02-24 | 2016-04-26 | Matterport, Inc. | Capturing and aligning three-dimensional scenes |
US11094137B2 (en) | 2012-02-24 | 2021-08-17 | Matterport, Inc. | Employing three-dimensional (3D) data predicted from two-dimensional (2D) images using neural networks for 3D modeling applications and other applications |
US10848731B2 (en) | 2012-02-24 | 2020-11-24 | Matterport, Inc. | Capturing and aligning panoramic image and depth data |
US10176589B2 (en) * | 2017-01-31 | 2019-01-08 | Mitsubishi Electric Research Labroatories, Inc. | Method and system for completing point clouds using planar segments |
US10735707B2 (en) | 2017-08-15 | 2020-08-04 | International Business Machines Corporation | Generating three-dimensional imagery |
WO2019107536A1 (en) * | 2017-11-30 | 2019-06-06 | 三菱電機株式会社 | Three-dimensional map generating system, three-dimensional map generating method, and three-dimensional map generating program |
KR102595787B1 (en) | 2018-02-27 | 2023-11-24 | 삼성전자주식회사 | Electronic device and control method thereof |
AT521845B1 (en) * | 2018-09-26 | 2021-05-15 | Waits Martin | Method for adjusting the focus of a film camera |
KR102526700B1 (en) | 2018-12-12 | 2023-04-28 | 삼성전자주식회사 | Electronic device and method for displaying three dimensions image |
CN113822924B (en) * | 2019-01-28 | 2024-04-05 | 弗劳恩霍夫应用研究促进协会 | Element localization in space |
CN111583328B (en) * | 2020-05-06 | 2021-10-22 | 南京农业大学 | Three-dimensional estimation method for epipremnum aureum leaf external phenotype parameters based on geometric model |
US11676406B2 (en) | 2020-05-20 | 2023-06-13 | Applications Mobiles Overview Inc. | System and method of augmenting a three-dimensional objects training dataset |
KR20220064045A (en) | 2020-11-11 | 2022-05-18 | 삼성전자주식회사 | Method and apparatus of generating image, and method of training artificial neural network for image generation |
Family Cites Families (390)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US2593925A (en) | 1948-10-05 | 1952-04-22 | Sheldon Edward Emanuel | Device for color projection of invisible rays |
US2804500A (en) | 1953-10-01 | 1957-08-27 | Rca Corp | Color interpretation system |
US2799722A (en) | 1954-04-26 | 1957-07-16 | Adalia Ltd | Reproduction with localized corrections |
US2874212A (en) | 1955-07-29 | 1959-02-17 | Rca Corp | Generator of color images from monochrome television signals |
US2883763A (en) | 1956-09-28 | 1959-04-28 | Otto F Schaper | Carrier landing trainer |
US2974190A (en) | 1957-12-09 | 1961-03-07 | Columbia Broadcasting Syst Inc | Electronic matting apparatus |
US3005042A (en) | 1958-04-17 | 1961-10-17 | David S Horsley | Electronic motion picture printer |
US3258528A (en) | 1963-06-18 | 1966-06-28 | Gen Precision Inc | Converter for changing a black and white television signal to a color television signal |
US3551589A (en) | 1967-03-23 | 1970-12-29 | Ward Electronic Ind | Apparatus for converting monochrome television signals to color signals |
US3558811A (en) | 1967-05-25 | 1971-01-26 | Xerox Corp | Graphic communication electrical interface system |
US3486242A (en) | 1967-05-29 | 1969-12-30 | Us Navy | Assault boat coxswain trainer |
US3560644A (en) | 1968-02-29 | 1971-02-02 | Us Navy | Multiple projection television system |
US3621127A (en) | 1969-02-13 | 1971-11-16 | Karl Hope | Synchronized stereoscopic system |
US3595987A (en) | 1969-02-20 | 1971-07-27 | Ass Motion Picture Tv Prod | Electronic composite photography |
US3617626A (en) | 1969-05-16 | 1971-11-02 | Technicolor | High-definition color picture editing and recording system |
US3612755A (en) | 1969-07-03 | 1971-10-12 | Dorothea Weitzner | Color pattern generator |
US3619051A (en) | 1969-10-23 | 1971-11-09 | Norman Wright Productions Inc | Production of color films from monochromatic film |
US3761607A (en) | 1969-11-03 | 1973-09-25 | Technicolor | Video monochrom to color conversion |
US3603962A (en) | 1970-03-18 | 1971-09-07 | Rca Corp | Color display for computer terminal |
US3647942A (en) | 1970-04-23 | 1972-03-07 | Eric J Siegel | Video color synthesizer |
US3731995A (en) | 1970-10-29 | 1973-05-08 | Instructional Dynamics | Method and apparatus for producing animated motion pictures |
US3710011A (en) | 1970-12-04 | 1973-01-09 | Computer Image Corp | System for automatically producing a color display of a scene from a black and white representation of the scene |
US3673317A (en) | 1970-12-30 | 1972-06-27 | Westinghouse Electric Corp | Comparitive display of images in color |
US3772465A (en) | 1971-06-09 | 1973-11-13 | Ass Of Motion Picture Televisi | Image modification of motion pictures |
US3742125A (en) | 1971-06-11 | 1973-06-26 | Electronic Visions Inc | Color video abstract synthesizer |
US3706841A (en) | 1971-09-17 | 1972-12-19 | Joseph F Novak | Method and apparatus for converting monochrome pictures to multi-color pictures electronically |
US3784736A (en) | 1971-09-17 | 1974-01-08 | J Novak | Method and apparatus for converting monochrome pictures to multi-color pictures electronically |
US3705762A (en) | 1971-09-20 | 1972-12-12 | Color Systems Inc | Method for converting black-and-white films to color films |
US3770885A (en) | 1971-10-21 | 1973-11-06 | Us Navy | Color electronic periscope view simulator |
US3737567A (en) | 1971-10-25 | 1973-06-05 | S Kratomi | Stereoscopic apparatus having liquid crystal filter viewer |
US3770884A (en) | 1972-02-07 | 1973-11-06 | Us Navy | Luminance control circuit for multi-color periscope view simulator |
US3769458A (en) | 1972-05-23 | 1973-10-30 | Us Navy | Color electronic synthesizer |
US4021846A (en) | 1972-09-25 | 1977-05-03 | The United States Of America As Represented By The Secretary Of The Navy | Liquid crystal stereoscopic viewer |
US3851955A (en) | 1973-02-05 | 1974-12-03 | Marks Polarized Corp | Apparatus for converting motion picture projectors for stereo display |
US4017166A (en) | 1973-02-05 | 1977-04-12 | Marks Polarized Corporation | Motion picture film for three dimensional projection |
US4183633A (en) | 1973-02-05 | 1980-01-15 | Marks Polarized Corporation | Motion picture film for three dimensional projection |
US3848856A (en) | 1973-10-01 | 1974-11-19 | Hazeltine Corp | Local correction apparatus for a color previewer |
US4168885A (en) | 1974-11-18 | 1979-09-25 | Marks Polarized Corporation | Compatible 3-dimensional motion picture projection system |
US3972067A (en) | 1975-01-17 | 1976-07-27 | The Singer Company | Color video synthesizer with monochrome input |
GB1547666A (en) | 1975-04-05 | 1979-06-27 | Nippon Electric Co | Video signal coding system |
US3971068A (en) | 1975-08-22 | 1976-07-20 | The United States Of America As Represented By The Secretary Of The Navy | Image processing system |
US4021841A (en) | 1975-12-31 | 1977-05-03 | Ralph Weinger | Color video synthesizer with improved image control means |
US4189743A (en) | 1976-12-20 | 1980-02-19 | New York Institute Of Technology | Apparatus and method for automatic coloration and/or shading of images |
US4189744A (en) | 1976-12-20 | 1980-02-19 | New York Institute Of Technology | Apparatus for generating signals representing operator-selected portions of a scene |
US4149185A (en) | 1977-03-04 | 1979-04-10 | Ralph Weinger | Apparatus and method for animated conversion of black and white video to color |
US4235503A (en) | 1978-05-08 | 1980-11-25 | Condon Chris J | Film projection lens system for 3-D movies |
US4183046A (en) | 1978-08-17 | 1980-01-08 | Interpretation Systems Incorporated | Electronic apparatus for converting digital image or graphics data to color video display formats and method therefor |
US4258385A (en) | 1979-05-15 | 1981-03-24 | Combined Logic Company | Interactive video production system and method |
DE3024459A1 (en) | 1979-07-03 | 1981-01-08 | Crosfield Electronics Ltd | PYRAMID INTERPOLATION |
EP0024862A3 (en) | 1979-09-04 | 1981-03-25 | Harold Charles Taylor | Video apparatus for visualing effects of selected juxtaposed colours |
US4318121A (en) | 1980-05-06 | 1982-03-02 | Jason Taite | Interior decor composition and display systems |
US4436369A (en) | 1981-09-08 | 1984-03-13 | Optimax Iii, Inc. | Stereoscopic lens system |
US4617592A (en) | 1982-03-11 | 1986-10-14 | Crosfield Electronics Limited | Video retouching system |
GB8306339D0 (en) | 1982-03-19 | 1983-04-13 | Quantel Ltd | Video processing systems |
US4549172A (en) | 1982-06-21 | 1985-10-22 | Motorola, Inc. | Multicolor display from monochrome or multicolor control unit |
US4645459A (en) | 1982-07-30 | 1987-02-24 | Honeywell Inc. | Computer generated synthesized imagery |
US4600919A (en) | 1982-08-03 | 1986-07-15 | New York Institute Of Technology | Three dimensional animation |
SU1192168A1 (en) | 1982-11-09 | 1985-11-15 | Vladimir A Gornykh | Method and apparatus for generating and reproducing television signal of pseudostereoscopic picture |
JPS59116736A (en) | 1982-12-24 | 1984-07-05 | Fuotoron:Kk | Stereoscopic projector |
US4475104A (en) | 1983-01-17 | 1984-10-02 | Lexidata Corporation | Three-dimensional display system |
US4603952A (en) | 1983-04-18 | 1986-08-05 | Sybenga John R | Professional stereoscopic projection |
CA1258118A (en) | 1983-05-05 | 1989-08-01 | Wilson Markle | Method of, and apparatus for, colouring a black and white video signal |
US4925294A (en) | 1986-12-17 | 1990-05-15 | Geshwind David M | Method to convert two dimensional motion pictures for three-dimensional systems |
US5050984A (en) | 1983-05-09 | 1991-09-24 | Geshwind David M | Method for colorizing footage |
US6590573B1 (en) | 1983-05-09 | 2003-07-08 | David Michael Geshwind | Interactive computer system for creating three-dimensional image information and for converting two-dimensional image information for three-dimensional display systems |
US4606625A (en) | 1983-05-09 | 1986-08-19 | Geshwind David M | Method for colorizing black and white footage |
US4755870A (en) | 1983-07-11 | 1988-07-05 | Colorization Inc. | Coloring a black and white signal using motion detection |
JPS6052190A (en) | 1983-08-31 | 1985-03-25 | Daitsuu:Kk | Video processing device |
US4608596A (en) | 1983-09-09 | 1986-08-26 | New York Institute Of Technology | System for colorizing video with both pseudo-colors and selected colors |
US4700181A (en) | 1983-09-30 | 1987-10-13 | Computer Graphics Laboratories, Inc. | Graphics display system |
US4558359A (en) | 1983-11-01 | 1985-12-10 | The United States Of America As Represented By The Secretary Of The Air Force | Anaglyphic stereoscopic image apparatus and method |
US4647965A (en) | 1983-11-02 | 1987-03-03 | Imsand Donald J | Picture processing system for three dimensional movies and video systems |
US4723159A (en) | 1983-11-02 | 1988-02-02 | Imsand Donald J | Three dimensional television and video systems |
US4590511A (en) | 1984-01-03 | 1986-05-20 | Honeywell Inc. | Circuit for converting the phase encoded hue information of a quadrature modulated color subcarrier into distinct analog voltage levels |
GB8405947D0 (en) | 1984-03-07 | 1984-04-11 | Quantel Ltd | Video signal processing systems |
US4694329A (en) | 1984-04-09 | 1987-09-15 | Corporate Communications Consultants, Inc. | Color correction system and method with scene-change detection |
US4721951A (en) | 1984-04-27 | 1988-01-26 | Ampex Corporation | Method and apparatus for color selection and production |
US4697178A (en) | 1984-06-29 | 1987-09-29 | Megatek Corporation | Computer graphics system for real-time calculation and display of the perspective view of three-dimensional scenes |
GB8422209D0 (en) | 1984-09-03 | 1984-10-10 | Crosfield Electronics Ltd | Image retouching |
US4642676A (en) | 1984-09-10 | 1987-02-10 | Color Systems Technology, Inc. | Priority masking techniques for video special effects |
DE3444353A1 (en) | 1984-12-05 | 1986-06-12 | Hasso 2000 Hamburg Hofmann | Method and device for generating images, particularly for video games and for producing graphical representations |
US4760390A (en) | 1985-02-25 | 1988-07-26 | Computer Graphics Laboratories, Inc. | Graphics display system and method with enhanced instruction data and processing |
JPH0681275B2 (en) | 1985-04-03 | 1994-10-12 | ソニー株式会社 | Image converter |
US4827255A (en) | 1985-05-31 | 1989-05-02 | Ascii Corporation | Display control system which produces varying patterns to reduce flickering |
JPS62216478A (en) | 1986-03-17 | 1987-09-24 | Dainippon Screen Mfg Co Ltd | Tint laying device |
US4888713B1 (en) | 1986-09-05 | 1993-10-12 | Cdi Technologies, Inc. | Surface detail mapping system |
US4758908A (en) | 1986-09-12 | 1988-07-19 | Fred James | Method and apparatus for substituting a higher quality audio soundtrack for a lesser quality audio soundtrack during reproduction of the lesser quality audio soundtrack and a corresponding visual picture |
US4809065A (en) | 1986-12-01 | 1989-02-28 | Kabushiki Kaisha Toshiba | Interactive system and related method for displaying data to produce a three-dimensional image of an object |
US4725879A (en) | 1987-03-27 | 1988-02-16 | Honeywell Inc. | Chroma responsive inspection apparatus selectively producing analog voltage levels based on the luminance, the phase of the chrominance subcarrier, or the amplitude of the chrominance subcarrier |
JPH03501194A (en) | 1987-07-27 | 1991-03-14 | ゲシュウィンド、デイビッド、エム | How to transmit high-definition television through narrow bandwidth channels |
US5093717A (en) | 1987-08-03 | 1992-03-03 | American Film Technologies, Inc. | System and method for digitally coloring images |
US4984072A (en) | 1987-08-03 | 1991-01-08 | American Film Technologies, Inc. | System and method for color image enhancement |
ES2083360T3 (en) | 1987-08-03 | 1996-04-16 | American Film Tech | SYSTEM AND METHOD FOR INTENSIFYING THE COLOR OF AN IMAGE. |
DE3736790A1 (en) | 1987-10-30 | 1989-05-11 | Broadcast Television Syst | METHOD FOR AUTOMATICALLY CORRECTING IMAGE ERRORS IN FILM SCANNING |
US5055939A (en) | 1987-12-15 | 1991-10-08 | Karamon John J | Method system & apparatus for synchronizing an auxiliary sound source containing multiple language channels with motion picture film video tape or other picture source containing a sound track |
US4918624A (en) | 1988-02-05 | 1990-04-17 | The United States Of America As Represented By The United States Department Of Energy | Vector generator scan converter |
US4933670A (en) | 1988-07-21 | 1990-06-12 | Picker International, Inc. | Multi-axis trackball |
US4952051A (en) | 1988-09-27 | 1990-08-28 | Lovell Douglas C | Method and apparatus for producing animated drawings and in-between drawings |
US5177474A (en) | 1989-09-13 | 1993-01-05 | Matsushita Electric Industrial Co., Ltd. | Three-dimensional display apparatus |
US5237647A (en) | 1989-09-15 | 1993-08-17 | Massachusetts Institute Of Technology | Computer aided drawing in three dimensions |
US5038161A (en) | 1990-01-08 | 1991-08-06 | Ki Lui S | Method and a camera for combination pictures in a photograph |
JP2621568B2 (en) | 1990-01-11 | 1997-06-18 | ダイキン工業株式会社 | Graphic drawing method and apparatus |
US5428721A (en) | 1990-02-07 | 1995-06-27 | Kabushiki Kaisha Toshiba | Data processing apparatus for editing image by using image conversion |
US5002387A (en) | 1990-03-23 | 1991-03-26 | Imax Systems Corporation | Projection synchronization system |
US5252953A (en) | 1990-05-22 | 1993-10-12 | American Film Technologies, Inc. | Computergraphic animation system |
US5181181A (en) | 1990-09-27 | 1993-01-19 | Triton Technologies, Inc. | Computer apparatus input device for three-dimensional information |
US5481321A (en) | 1991-01-29 | 1996-01-02 | Stereographics Corp. | Stereoscopic motion picture projection system |
US5185852A (en) | 1991-05-31 | 1993-02-09 | Digital Equipment Corporation | Antialiasing apparatus and method for computer printers |
US5347620A (en) | 1991-09-05 | 1994-09-13 | Zimmer Mark A | System and method for digital rendering of images and printed articulation |
AU650179B2 (en) | 1991-10-07 | 1994-06-09 | Eastman Kodak Company | A compositer interface for arranging the components of special effects for a motion picture production |
US5243460A (en) | 1991-11-08 | 1993-09-07 | Elliot Kornberg | Light filtering system for creating perception of stereoscopic images from two-dimensional images and enhancing perception of objects |
US7006881B1 (en) | 1991-12-23 | 2006-02-28 | Steven Hoffberg | Media recording device with remote graphic user interface |
US6850252B1 (en) | 1999-10-05 | 2005-02-01 | Steven M. Hoffberg | Intelligent electronic appliance system and method |
JP3117097B2 (en) | 1992-01-28 | 2000-12-11 | ソニー株式会社 | Image conversion device |
US5262856A (en) | 1992-06-04 | 1993-11-16 | Massachusetts Institute Of Technology | Video image compositing techniques |
US5328073A (en) | 1992-06-24 | 1994-07-12 | Eastman Kodak Company | Film registration and ironing gate assembly |
US5973700A (en) | 1992-09-16 | 1999-10-26 | Eastman Kodak Company | Method and apparatus for optimizing the resolution of images which have an apparent depth |
US5534915A (en) | 1992-09-30 | 1996-07-09 | American Film Technologies, Inc. | Method of color enhancing a monochrome image using multiple base colors for selected regions of the monochrome image |
US6011581A (en) | 1992-11-16 | 2000-01-04 | Reveo, Inc. | Intelligent method and system for producing and displaying stereoscopically-multiplexed images of three-dimensional objects for use in realistic stereoscopic viewing thereof in interactive virtual reality display environments |
US5734915A (en) | 1992-11-25 | 1998-03-31 | Eastman Kodak Company | Method and apparatus for composing digital medical imagery |
US5402191A (en) | 1992-12-09 | 1995-03-28 | Imax Corporation | Method and apparatus for presenting stereoscopic images |
US5495576A (en) | 1993-01-11 | 1996-02-27 | Ritchey; Kurtis J. | Panoramic image based virtual reality/telepresence audio-visual system and method |
JP3230618B2 (en) | 1993-03-05 | 2001-11-19 | 株式会社日立メディコ | Radiation stereoscopic imaging equipment |
US5717454A (en) | 1993-07-14 | 1998-02-10 | Lifetouch Portrait Studios, Inc. | Method and apparatus for creating posing masks on video screen |
US5739844A (en) | 1994-02-04 | 1998-04-14 | Sanyo Electric Co. Ltd. | Method of converting two-dimensional image into three-dimensional image |
US6553184B1 (en) | 1994-03-14 | 2003-04-22 | Sanyo Electric Co., Ltd. | Method of converting two dimensional images into three-dimensional images |
JP3486461B2 (en) | 1994-06-24 | 2004-01-13 | キヤノン株式会社 | Image processing apparatus and method |
JP2846840B2 (en) | 1994-07-14 | 1999-01-13 | 三洋電機株式会社 | Method for generating 3D image from 2D image |
AUPM701394A0 (en) | 1994-07-22 | 1994-08-18 | Monash University | A graphical display system |
JP2951230B2 (en) | 1994-09-22 | 1999-09-20 | 三洋電機株式会社 | Method for generating 3D image from 2D image |
DE69528946T2 (en) | 1994-09-22 | 2003-10-02 | Sanyo Electric Co | Process for converting two-dimensional images into three-dimensional images |
US5668605A (en) | 1994-10-25 | 1997-09-16 | R. T. Set | Object keying in video images based on distance from camera |
US5649032A (en) | 1994-11-14 | 1997-07-15 | David Sarnoff Research Center, Inc. | System for automatically aligning images to form a mosaic image |
JP2846830B2 (en) | 1994-11-22 | 1999-01-13 | 三洋電機株式会社 | How to convert 2D video to 3D video |
JPH08186844A (en) | 1994-12-28 | 1996-07-16 | Sanyo Electric Co Ltd | Stereoscopic picture generator and stereoscopic picture generating method |
JP3568606B2 (en) | 1995-01-17 | 2004-09-22 | 富士写真フイルム株式会社 | Image noise reduction processing method |
JP3265893B2 (en) | 1995-02-13 | 2002-03-18 | 株式会社日立製作所 | Image display device |
US5729471A (en) | 1995-03-31 | 1998-03-17 | The Regents Of The University Of California | Machine dynamic selection of one video camera/image of a scene from multiple video cameras/images of the scene in accordance with a particular perspective on the scene, an object in the scene, or an event in the scene |
US5899861A (en) | 1995-03-31 | 1999-05-04 | Siemens Medical Systems, Inc. | 3-dimensional volume by aggregating ultrasound fields of view |
US5699444A (en) | 1995-03-31 | 1997-12-16 | Synthonics Incorporated | Methods and apparatus for using image data to determine camera location and orientation |
US5777666A (en) | 1995-04-17 | 1998-07-07 | Sanyo Electric Co., Ltd. | Method of converting two-dimensional images into three-dimensional images |
US5742291A (en) | 1995-05-09 | 1998-04-21 | Synthonics Incorporated | Method and apparatus for creation of three-dimensional wire frames |
JPH08331473A (en) | 1995-05-29 | 1996-12-13 | Hitachi Ltd | Display device for television signal |
US5684715A (en) | 1995-06-07 | 1997-11-04 | Canon Information Systems, Inc. | Interactive video system with dynamic video object descriptors |
GB9513658D0 (en) | 1995-07-05 | 1995-09-06 | Philips Electronics Uk Ltd | Autostereoscopic display apparatus |
US5999660A (en) | 1995-07-26 | 1999-12-07 | California Institute Of Technology | Imaging system for correction of perceptual distortion in wide angle images |
JP2994232B2 (en) | 1995-07-28 | 1999-12-27 | ウシオ電機株式会社 | Mask-to-mask or mask-to-work alignment method and apparatus |
US6005582A (en) | 1995-08-04 | 1999-12-21 | Microsoft Corporation | Method and system for texture mapping images with anisotropic filtering |
US6049628A (en) | 1995-09-01 | 2000-04-11 | Cerulean Colorization Llc | Polygon reshaping in picture colorization |
US6263101B1 (en) | 1995-09-01 | 2001-07-17 | Cerulean Colorization Llc | Filtering in picture colorization |
US5825929A (en) | 1995-10-05 | 1998-10-20 | Microsoft Corporation | Transformation block optimization method |
US5897413A (en) | 1995-10-10 | 1999-04-27 | Erland; Jonathan | Travelling mat backing |
US5912994A (en) | 1995-10-27 | 1999-06-15 | Cerulean Colorization Llc | Methods for defining mask of substantially color-homogeneous regions of digitized picture stock |
EP0809913B1 (en) | 1995-12-19 | 2002-06-12 | Koninklijke Philips Electronics N.V. | Parallactic depth-dependent pixel shifts |
US6088006A (en) | 1995-12-20 | 2000-07-11 | Olympus Optical Co., Ltd. | Stereoscopic image generating system for substantially matching visual range with vergence distance |
US5748199A (en) | 1995-12-20 | 1998-05-05 | Synthonics Incorporated | Method and apparatus for converting a two dimensional motion picture into a three dimensional motion picture |
US5835163A (en) | 1995-12-21 | 1998-11-10 | Siemens Corporate Research, Inc. | Apparatus for detecting a cut in a video |
AUPN732395A0 (en) | 1995-12-22 | 1996-01-25 | Xenotech Research Pty Ltd | Image conversion and encoding techniques |
US5973831A (en) | 1996-01-22 | 1999-10-26 | Kleinberger; Paul | Systems for three-dimensional viewing using light polarizing layers |
US5841512A (en) | 1996-02-27 | 1998-11-24 | Goodhill; Dean Kenneth | Methods of previewing and editing motion pictures |
AU1983397A (en) | 1996-02-29 | 1997-09-16 | Acuson Corporation | Multiple ultrasound image registration system, method and transducer |
US5880788A (en) | 1996-03-25 | 1999-03-09 | Interval Research Corporation | Automated synchronization of video image sequences to new soundtracks |
US5867169A (en) | 1996-04-17 | 1999-02-02 | Pixar | Method and apparatus for manipulating color values in a computer graphics system |
BR9709751A (en) | 1996-04-29 | 1999-08-10 | Princeton Video Image Inc | Enhanced electronic insertion of video signals audio |
JP3258236B2 (en) | 1996-05-28 | 2002-02-18 | 株式会社日立製作所 | Multimedia information transfer system |
US6329963B1 (en) | 1996-06-05 | 2001-12-11 | Cyberlogic, Inc. | Three-dimensional display system: apparatus and method |
US5778108A (en) | 1996-06-07 | 1998-07-07 | Electronic Data Systems Corporation | Method and system for detecting transitional markers such as uniform fields in a video signal |
US5959697A (en) | 1996-06-07 | 1999-09-28 | Electronic Data Systems Corporation | Method and system for detecting dissolve transitions in a video signal |
US5767923A (en) | 1996-06-07 | 1998-06-16 | Electronic Data Systems Corporation | Method and system for detecting cuts in a video signal |
US5920360A (en) | 1996-06-07 | 1999-07-06 | Electronic Data Systems Corporation | Method and system for detecting fade transitions in a video signal |
US6198484B1 (en) | 1996-06-27 | 2001-03-06 | Kabushiki Kaisha Toshiba | Stereoscopic display system |
US6061067A (en) | 1996-08-02 | 2000-05-09 | Autodesk, Inc. | Applying modifiers to objects based on the types of the objects |
US6108005A (en) | 1996-08-30 | 2000-08-22 | Space Corporation | Method for producing a synthesized stereoscopic image |
GB9619119D0 (en) | 1996-09-12 | 1996-10-23 | Discreet Logic Inc | Processing image |
US20030093790A1 (en) | 2000-03-28 | 2003-05-15 | Logan James D. | Audio and video program recording, editing and playback systems using metadata |
WO1998027516A1 (en) | 1996-12-19 | 1998-06-25 | Koninklijke Philips Electronics N.V. | Method and device for displaying an autostereogram |
JPH10191020A (en) | 1996-12-20 | 1998-07-21 | Canon Inc | Object image segmenting method and device |
US5990903A (en) | 1997-02-03 | 1999-11-23 | Micron Technologies, Inc. | Method and apparatus for performing chroma key, transparency and fog operations |
US6727938B1 (en) | 1997-04-14 | 2004-04-27 | Robert Bosch Gmbh | Security system with maskable motion detection and camera with an adjustable field of view |
US6067125A (en) | 1997-05-15 | 2000-05-23 | Minerva Systems | Structure and method for film grain noise reduction |
US6492986B1 (en) | 1997-06-02 | 2002-12-10 | The Trustees Of The University Of Pennsylvania | Method for human face shape and motion estimation based on integrating optical flow and deformable models |
US6141433A (en) | 1997-06-19 | 2000-10-31 | Ncr Corporation | System and method for segmenting image regions from a scene likely to represent particular objects in the scene |
US6031564A (en) | 1997-07-07 | 2000-02-29 | Reveo, Inc. | Method and apparatus for monoscopic to stereoscopic image conversion |
US6157747A (en) | 1997-08-01 | 2000-12-05 | Microsoft Corporation | 3-dimensional image rotation method and apparatus for producing image mosaics |
AUPO894497A0 (en) | 1997-09-02 | 1997-09-25 | Xenotech Research Pty Ltd | Image processing method and apparatus |
US6414678B1 (en) | 1997-11-20 | 2002-07-02 | Nintendo Co., Ltd. | Image creating apparatus and image display apparatus |
AUPP048097A0 (en) | 1997-11-21 | 1997-12-18 | Xenotech Research Pty Ltd | Eye tracking apparatus |
US6166744A (en) | 1997-11-26 | 2000-12-26 | Pathfinder Systems, Inc. | System for combining virtual images with real-world scenes |
US6119123A (en) | 1997-12-02 | 2000-09-12 | U.S. Philips Corporation | Apparatus and method for optimizing keyframe and blob retrieval and storage |
ID27878A (en) | 1997-12-05 | 2001-05-03 | Dynamic Digital Depth Res Pty | IMAGE IMPROVED IMAGE CONVERSION AND ENCODING ENGINEERING |
US5990900A (en) | 1997-12-24 | 1999-11-23 | Be There Now, Inc. | Two-dimensional to three-dimensional image converting system |
JP4056154B2 (en) | 1997-12-30 | 2008-03-05 | 三星電子株式会社 | 2D continuous video 3D video conversion apparatus and method, and 3D video post-processing method |
US6973434B2 (en) | 1998-01-09 | 2005-12-06 | Millermed Software, Inc. | Computer-based system for automating administrative procedures in an office |
US6226015B1 (en) | 1998-02-25 | 2001-05-01 | Intel Corporation | Method of automatically producing sketches and cartoon images from movies |
US6847737B1 (en) | 1998-03-13 | 2005-01-25 | University Of Houston System | Methods for performing DAF data filtering and padding |
US6271859B1 (en) | 1998-04-06 | 2001-08-07 | Adobe Systems Incorporated | Recoloring art work |
US6677944B1 (en) | 1998-04-14 | 2004-01-13 | Shima Seiki Manufacturing Limited | Three-dimensional image generating apparatus that creates a three-dimensional model from a two-dimensional image by image processing |
US6363170B1 (en) | 1998-04-30 | 2002-03-26 | Wisconsin Alumni Research Foundation | Photorealistic scene reconstruction by voxel coloring |
US6957341B2 (en) | 1998-05-14 | 2005-10-18 | Purdue Research Foundation | Method and system for secure computational outsourcing and disguise |
US20050231505A1 (en) | 1998-05-27 | 2005-10-20 | Kaye Michael C | Method for creating artifact free three-dimensional images converted from two-dimensional images |
US7102633B2 (en) | 1998-05-27 | 2006-09-05 | In-Three, Inc. | Method for conforming objects to a common depth perspective for converting two-dimensional images into three-dimensional images |
US6515659B1 (en) | 1998-05-27 | 2003-02-04 | In-Three, Inc. | Method and system for creating realistic smooth three-dimensional depth contours from two-dimensional images |
US20050146521A1 (en) | 1998-05-27 | 2005-07-07 | Kaye Michael C. | Method for creating and presenting an accurate reproduction of three-dimensional images converted from two-dimensional images |
US7116324B2 (en) | 1998-05-27 | 2006-10-03 | In-Three, Inc. | Method for minimizing visual artifacts converting two-dimensional motion pictures into three-dimensional motion pictures |
US7116323B2 (en) | 1998-05-27 | 2006-10-03 | In-Three, Inc. | Method of hidden surface reconstruction for creating accurate three-dimensional images converted from two-dimensional images |
US6208348B1 (en) | 1998-05-27 | 2001-03-27 | In-Three, Inc. | System and method for dimensionalization processing of images in consideration of a pedetermined image projection format |
US6056691A (en) | 1998-06-24 | 2000-05-02 | Ecton, Inc. | System for collecting ultrasound imaging data at an adjustable collection image frame rate |
US6426750B1 (en) | 1998-07-14 | 2002-07-30 | Microsoft Corporation | Run-time geomorphs |
US20020001045A1 (en) | 1998-07-16 | 2002-01-03 | Minakanagurki Ranganath | Parallax viewing system |
US6456340B1 (en) | 1998-08-12 | 2002-09-24 | Pixonics, Llc | Apparatus and method for performing image transforms in a digital display system |
US6480822B2 (en) | 1998-08-24 | 2002-11-12 | Conexant Systems, Inc. | Low complexity random codebook structure |
US6466205B2 (en) | 1998-11-19 | 2002-10-15 | Push Entertainment, Inc. | System and method for creating 3D models from 2D sequential image data |
GB2344037B (en) | 1998-11-20 | 2003-01-22 | Ibm | A method and apparatus for adjusting the display scale of an image |
US6707487B1 (en) | 1998-11-20 | 2004-03-16 | In The Play, Inc. | Method for representing real-time motion |
US6364835B1 (en) | 1998-11-20 | 2002-04-02 | Acuson Corporation | Medical diagnostic ultrasound imaging methods for extended field of view |
AUPP727598A0 (en) | 1998-11-23 | 1998-12-17 | Dynamic Digital Depth Research Pty Ltd | Improved teleconferencing system |
US6390980B1 (en) | 1998-12-07 | 2002-05-21 | Atl Ultrasound, Inc. | Spatial compounding with ultrasonic doppler signal information |
US6373970B1 (en) | 1998-12-29 | 2002-04-16 | General Electric Company | Image registration using fourier phase matching |
US6429867B1 (en) | 1999-03-15 | 2002-08-06 | Sun Microsystems, Inc. | System and method for generating and playback of three-dimensional movies |
US6606166B1 (en) | 1999-04-30 | 2003-08-12 | Adobe Systems Incorporated | Pattern dithering |
ES2205752T5 (en) | 1999-05-24 | 2007-12-01 | Pilkington Italia S.P.A. | MOLDED OF AN ELASTOMERIC PROFILE OF A GLASS. |
US6405366B1 (en) | 1999-05-28 | 2002-06-11 | Electronic Data Systems Corporation | Multi-layered software application interface architecture |
AUPQ101899A0 (en) | 1999-06-17 | 1999-07-08 | Dynamic Digital Depth Research Pty Ltd | Image enhancement system |
AUPQ119799A0 (en) | 1999-06-25 | 1999-07-22 | Dynamic Digital Depth Research Pty Ltd | Ddc/3 image conversion and encoding techniques |
US6662357B1 (en) | 1999-08-31 | 2003-12-09 | Accenture Llp | Managing information in an integrated development architecture framework |
JP3722653B2 (en) | 1999-08-31 | 2005-11-30 | 松下電器産業株式会社 | Surveillance camera device and display method of surveillance camera |
GB2354389A (en) | 1999-09-15 | 2001-03-21 | Sharp Kk | Stereo images with comfortable perceived depth |
US6964009B2 (en) | 1999-10-21 | 2005-11-08 | Automated Media Processing Solutions, Inc. | Automated media delivery system |
US20010025267A1 (en) | 2000-01-14 | 2001-09-27 | Stephen Janiszewski | System and method for facilitating bidding transactions and conducting project management utilizing software metric collection |
US20030069777A1 (en) | 2000-01-31 | 2003-04-10 | Zvi Or-Bach | Integrated system for providing remote outsourcing of services |
AU2001241475A1 (en) | 2000-02-11 | 2001-08-20 | Comsat Corporation | Background noise reduction in sinusoidal based speech coding systems |
US6737957B1 (en) | 2000-02-16 | 2004-05-18 | Verance Corporation | Remote control signaling using audio watermarks |
US6509926B1 (en) | 2000-02-17 | 2003-01-21 | Sensormatic Electronics Corporation | Surveillance apparatus for camera surveillance system |
US6807290B2 (en) | 2000-03-09 | 2004-10-19 | Microsoft Corporation | Rapid computer modeling of faces for animation |
US20020049778A1 (en) | 2000-03-31 | 2002-04-25 | Bell Peter W. | System and method of information outsourcing |
DE10016074B4 (en) | 2000-04-01 | 2004-09-30 | Tdv Technologies Corp. | Method and device for generating 3D images |
US7084868B2 (en) | 2000-04-26 | 2006-08-01 | University Of Louisville Research Foundation, Inc. | System and method for 3-D digital reconstruction of an oral cavity from a sequence of 2-D images |
US6665798B1 (en) | 2000-04-27 | 2003-12-16 | International Business Machines Corporation | System for assigning data processing activities to users through an interactive display interface dynamically granting access only during activity to normally inaccessible resources needed for activity |
US6611268B1 (en) | 2000-05-30 | 2003-08-26 | Microsoft Corporation | System and process for generating 3D video textures using video-based rendering techniques |
US20010051913A1 (en) | 2000-06-07 | 2001-12-13 | Avinash Vashistha | Method and system for outsourcing information technology projects and services |
AU2001266862A1 (en) | 2000-06-12 | 2001-12-24 | Vrex, Inc. | Electronic stereoscopic media delivery system |
WO2002008948A2 (en) | 2000-07-24 | 2002-01-31 | Vivcom, Inc. | System and method for indexing, searching, identifying, and editing portions of electronic multimedia files |
KR20030029649A (en) | 2000-08-04 | 2003-04-14 | 다이나믹 디지탈 텝스 리서치 피티와이 엘티디 | Image conversion and encoding technique |
CA2418800A1 (en) | 2000-08-09 | 2002-02-14 | Dynamic Digital Depth Research Pty Ltd. | Image conversion and encoding techniques |
JP3819841B2 (en) | 2000-08-18 | 2006-09-13 | 株式会社アマダ | Outsourcing service equipment for electronic drawing data |
US6416477B1 (en) | 2000-08-22 | 2002-07-09 | Koninklijke Philips Electronics N.V. | Ultrasonic diagnostic systems with spatial compounded panoramic imaging |
US7000223B1 (en) | 2000-09-08 | 2006-02-14 | Corel Corporation | Method and apparatus for preparing a definition to control automated data processing |
DE10193433T1 (en) | 2000-09-18 | 2002-10-31 | Holon Co Ltd | Mask inspection device |
JP2002123842A (en) | 2000-10-13 | 2002-04-26 | Takumi:Kk | Device for generating stereoscopic image, and medium for recording information |
ATE552572T1 (en) | 2000-12-01 | 2012-04-15 | Imax Corp | METHOD AND DEVICE FOR GENERATING HIGH-RESOLUTION IMAGES |
US7117231B2 (en) | 2000-12-07 | 2006-10-03 | International Business Machines Corporation | Method and system for the automatic generation of multi-lingual synchronized sub-titles for audiovisual data |
FR2820269A1 (en) | 2001-01-30 | 2002-08-02 | Koninkl Philips Electronics Nv | PROCESS FOR PROCESSING 2D IMAGES APPLIED TO 3D OBJECTS |
KR100360893B1 (en) | 2001-02-01 | 2002-11-13 | 엘지전자 주식회사 | Apparatus and method for compensating video motions |
US7461002B2 (en) | 2001-04-13 | 2008-12-02 | Dolby Laboratories Licensing Corporation | Method for time aligning audio signals using characterizations based on auditory events |
US8401336B2 (en) | 2001-05-04 | 2013-03-19 | Legend3D, Inc. | System and method for rapid image sequence depth enhancement with augmented computer-generated elements |
MXPA03010039A (en) | 2001-05-04 | 2004-12-06 | Legend Films Llc | Image sequence enhancement system and method. |
US6965379B2 (en) | 2001-05-08 | 2005-11-15 | Koninklijke Philips Electronics N.V. | N-view synthesis from monocular video of certain broadcast and stored mass media content |
JP2003046982A (en) | 2001-07-30 | 2003-02-14 | Wowow Inc | System for participating in cheer in tv broadcast |
US7123263B2 (en) | 2001-08-14 | 2006-10-17 | Pulse Entertainment, Inc. | Automatic 3D modeling system and method |
US7343082B2 (en) | 2001-09-12 | 2008-03-11 | Ryshco Media Inc. | Universal guide track |
US20030097423A1 (en) | 2001-10-26 | 2003-05-22 | Yuka Ozawa | Preview system for data broadcast contents |
US20050188297A1 (en) | 2001-11-01 | 2005-08-25 | Automatic E-Learning, Llc | Multi-audio add/drop deterministic animation synchronization |
US6859523B1 (en) | 2001-11-14 | 2005-02-22 | Qgenisys, Inc. | Universal task management system, method and product for automatically managing remote workers, including assessing the work product and workers |
EP1451775A1 (en) | 2001-11-24 | 2004-09-01 | TDV Technologies Corp. | Generation of a stereo image sequence from a 2d image sequence |
US7032177B2 (en) | 2001-12-27 | 2006-04-18 | Digeo, Inc. | Method and system for distributing personalized editions of media programs using bookmarks |
JP3673217B2 (en) | 2001-12-20 | 2005-07-20 | オリンパス株式会社 | Video display device |
US8150235B2 (en) | 2002-02-08 | 2012-04-03 | Intel Corporation | Method of home media server control |
US7333519B2 (en) | 2002-04-23 | 2008-02-19 | Gateway Inc. | Method of manually fine tuning audio synchronization of a home network |
US6791542B2 (en) | 2002-06-17 | 2004-09-14 | Mitsubishi Electric Research Laboratories, Inc. | Modeling 3D objects with opacity hulls |
JP2004040445A (en) | 2002-07-03 | 2004-02-05 | Sharp Corp | Portable equipment having 3d display function and 3d transformation program |
US7308139B2 (en) | 2002-07-12 | 2007-12-11 | Chroma Energy, Inc. | Method, system, and apparatus for color representation of seismic data and associated measurements |
US8788650B1 (en) | 2002-07-19 | 2014-07-22 | Fortinet, Inc. | Hardware based detection devices for detecting network traffic content and methods of using the same |
US6919892B1 (en) | 2002-08-14 | 2005-07-19 | Avaworks, Incorporated | Photo realistic talking head creation system and method |
US7027054B1 (en) | 2002-08-14 | 2006-04-11 | Avaworks, Incorporated | Do-it-yourself photo realistic talking head creation system and method |
WO2004019621A1 (en) | 2002-08-20 | 2004-03-04 | Kazunari Era | Method and device for creating 3-dimensional view image |
US20030046656A1 (en) | 2002-08-29 | 2003-03-06 | Asha Saxena | Information technology outsourcing hubs |
US7024054B2 (en) | 2002-09-27 | 2006-04-04 | Eastman Kodak Company | Method and system for generating a foreground mask for a composite image |
US8108488B2 (en) | 2002-11-18 | 2012-01-31 | Jackbe Corporation | System and method for reducing bandwidth requirements for remote applications by utilizing client processing power |
JP2004207985A (en) | 2002-12-25 | 2004-07-22 | Fuji Photo Film Co Ltd | Digital camera |
KR100505334B1 (en) | 2003-03-28 | 2005-08-04 | (주)플렛디스 | Real-time stereoscopic image conversion apparatus using motion parallaxr |
US7098910B2 (en) | 2003-05-14 | 2006-08-29 | Lena Petrovic | Hair rendering method and apparatus |
WO2005010623A2 (en) | 2003-07-24 | 2005-02-03 | Zebra Imaging, Inc. | Enhanced environment visualization using holographic stereograms |
US7576332B2 (en) | 2003-08-01 | 2009-08-18 | St. George's Healthcare Nhs Trust | Imaging system |
US7136075B1 (en) | 2003-08-26 | 2006-11-14 | Adobe Systems Incorporated | Compositing in the presence of different transparencies |
US7532225B2 (en) | 2003-09-18 | 2009-05-12 | Kabushiki Kaisha Toshiba | Three-dimensional image display device |
US8090402B1 (en) | 2003-09-26 | 2012-01-03 | Iwao Fujisaki | Communication device |
US20050088515A1 (en) | 2003-10-23 | 2005-04-28 | Geng Z. J. | Camera ring for three-dimensional (3D) surface imaging |
US7663689B2 (en) | 2004-01-16 | 2010-02-16 | Sony Computer Entertainment Inc. | Method and apparatus for optimizing capture device settings through depth information |
ATE382919T1 (en) | 2004-02-17 | 2008-01-15 | Koninkl Philips Electronics Nv | CREATE A DEPTH MAP |
US8634072B2 (en) | 2004-03-06 | 2014-01-21 | Michael Trainer | Methods and apparatus for determining characteristics of particles |
US20050209983A1 (en) | 2004-03-18 | 2005-09-22 | Macpherson Deborah L | Context driven topologies |
US7505035B2 (en) | 2004-04-19 | 2009-03-17 | Oki Semiconductor Co., Ltd. | Power-down circuit for a display device |
US20060143059A1 (en) | 2004-04-26 | 2006-06-29 | Barry Sandrew | System and method for minimizing transfer of motion picture data manipulated with outsourced labor |
US20040181444A1 (en) | 2004-04-26 | 2004-09-16 | Sandrew Barry B. | System and method for manipulating motion picture data using labor outsourcing |
US7349570B2 (en) | 2004-06-16 | 2008-03-25 | Chuan-Sheng Chen | Graphic image to 3D image conversion device |
KR100624431B1 (en) | 2004-08-03 | 2006-09-19 | 삼성전자주식회사 | Method and apparatus controlled convergence distance for observation of stereo image |
US7542034B2 (en) | 2004-09-23 | 2009-06-02 | Conversion Works, Inc. | System and method for processing video images |
US20080246836A1 (en) | 2004-09-23 | 2008-10-09 | Conversion Works, Inc. | System and method for processing video images for camera recreation |
US20080259073A1 (en) | 2004-09-23 | 2008-10-23 | Conversion Works, Inc. | System and method for processing video images |
US7330584B2 (en) | 2004-10-14 | 2008-02-12 | Sony Corporation | Image processing apparatus and method |
KR100603601B1 (en) | 2004-11-08 | 2006-07-24 | 한국전자통신연구원 | Apparatus and Method for Production Multi-view Contents |
CA2529112A1 (en) | 2004-12-28 | 2006-06-28 | Weyerhaeuser Company | Methods for processing image and/or spectral data for enhanced embryo classification |
US7639876B2 (en) | 2005-01-14 | 2009-12-29 | Advanced Digital Systems, Inc. | System and method for associating handwritten information with one or more objects |
WO2006078237A1 (en) | 2005-01-15 | 2006-07-27 | In-Three, Inc. | Method for converting two-dimensional images to three-dimensional images using depth perspective |
GB2422737A (en) | 2005-01-26 | 2006-08-02 | Sharp Kk | Multiple-view display and display controller |
KR20070119018A (en) | 2005-02-23 | 2007-12-18 | 크레이그 써머스 | Automatic scene modeling for the 3d camera and 3d video |
US7512262B2 (en) | 2005-02-25 | 2009-03-31 | Microsoft Corporation | Stereo-based image processing |
US8300841B2 (en) | 2005-06-03 | 2012-10-30 | Apple Inc. | Techniques for presenting sound effects on a portable media player |
US7079075B1 (en) | 2005-06-07 | 2006-07-18 | Trimble Navigation Limited | GPS rover station for synthesizing synthetic reference phases for controlling accuracy of high integrity positions |
US20070102622A1 (en) | 2005-07-01 | 2007-05-10 | Olsen Richard I | Apparatus for multiple camera devices and method of operating same |
US7860342B2 (en) | 2005-07-01 | 2010-12-28 | The Invention Science Fund I, Llc | Modifying restricted images |
CA2553473A1 (en) | 2005-07-26 | 2007-01-26 | Wa James Tam | Generating a depth map from a tw0-dimensional source image for stereoscopic and multiview imaging |
US7944454B2 (en) | 2005-09-07 | 2011-05-17 | Fuji Xerox Co., Ltd. | System and method for user monitoring interface of 3-D video streams from multiple cameras |
JP2007199684A (en) | 2005-12-28 | 2007-08-09 | Canon Inc | Image display apparatus |
US20070238981A1 (en) | 2006-03-13 | 2007-10-11 | Bracco Imaging Spa | Methods and apparatuses for recording and reviewing surgical navigation processes |
US20070236514A1 (en) | 2006-03-29 | 2007-10-11 | Bracco Imaging Spa | Methods and Apparatuses for Stereoscopic Image Guided Surgical Navigation |
US20070260634A1 (en) | 2006-05-04 | 2007-11-08 | Nokia Corporation | Apparatus, system, method, and computer program product for synchronizing the presentation of media content |
US7742642B2 (en) | 2006-05-30 | 2010-06-22 | Expedata, Llc | System and method for automated reading of handwriting |
US7573489B2 (en) | 2006-06-01 | 2009-08-11 | Industrial Light & Magic | Infilling for 2D to 3D image conversion |
US7573475B2 (en) | 2006-06-01 | 2009-08-11 | Industrial Light & Magic | 2D to 3D image conversion |
CN102685533B (en) | 2006-06-23 | 2015-03-18 | 图象公司 | Methods and systems for converting 2d motion pictures into stereoscopic 3d exhibition |
US20080002878A1 (en) | 2006-06-28 | 2008-01-03 | Somasundaram Meiyappan | Method For Fast Stereo Matching Of Images |
US7623755B2 (en) | 2006-08-17 | 2009-11-24 | Adobe Systems Incorporated | Techniques for positioning audio and video clips |
US7948558B2 (en) | 2006-09-29 | 2011-05-24 | The Directv Group, Inc. | Audio video timing measurement and synchronization |
WO2008075276A1 (en) | 2006-12-19 | 2008-06-26 | Koninklijke Philips Electronics N.V. | Method and system to convert 2d video into 3d video |
US7568057B2 (en) | 2006-12-19 | 2009-07-28 | Intel Corporation | Method and apparatus for maintaining synchronization of audio in a computing system |
JP5022025B2 (en) | 2006-12-27 | 2012-09-12 | インターナショナル・ビジネス・マシーンズ・コーポレーション | A method and apparatus for synchronizing content data streams and metadata. |
US8462109B2 (en) | 2007-01-05 | 2013-06-11 | Invensense, Inc. | Controlling and accessing content using motion processing on mobile devices |
CN101669143B (en) | 2007-01-11 | 2012-06-20 | Mvm知识产权有限公司 | The measurement of functional microcirculatory geometry and velocity distributions using automated image analysis |
US8462988B2 (en) | 2007-01-23 | 2013-06-11 | Valeo Schalter Und Sensoren Gmbh | Method and system for universal lane boundary detection |
US8655052B2 (en) | 2007-01-26 | 2014-02-18 | Intellectual Discovery Co., Ltd. | Methodology for 3D scene reconstruction from 2D image sequences |
US7665798B2 (en) | 2007-03-01 | 2010-02-23 | Chih-Yu Hsia | Means and methods to reduce drags |
US8274530B2 (en) | 2007-03-12 | 2012-09-25 | Conversion Works, Inc. | Systems and methods for filling occluded information for 2-D to 3-D conversion |
US20080225059A1 (en) | 2007-03-12 | 2008-09-18 | Conversion Works, Inc. | System and method for using off-screen mask space to provide enhanced viewing |
US20080225045A1 (en) | 2007-03-12 | 2008-09-18 | Conversion Works, Inc. | Systems and methods for 2-d to 3-d image conversion using mask to model, or model to mask, conversion |
US20080226128A1 (en) | 2007-03-12 | 2008-09-18 | Conversion Works, Inc. | System and method for using feature tracking techniques for the generation of masks in the conversion of two-dimensional images to three-dimensional images |
US20080226181A1 (en) | 2007-03-12 | 2008-09-18 | Conversion Works, Inc. | Systems and methods for depth peeling using stereoscopic variables during the rendering of 2-d to 3-d images |
US20080226194A1 (en) | 2007-03-12 | 2008-09-18 | Conversion Works, Inc. | Systems and methods for treating occlusions in 2-d to 3-d image conversion |
US20080228449A1 (en) | 2007-03-12 | 2008-09-18 | Conversion Works, Inc. | Systems and methods for 2-d to 3-d conversion using depth access segments to define an object |
US20080225040A1 (en) | 2007-03-12 | 2008-09-18 | Conversion Works, Inc. | System and method of treating semi-transparent features in the conversion of two-dimensional images to three-dimensional images |
US20080225042A1 (en) | 2007-03-12 | 2008-09-18 | Conversion Works, Inc. | Systems and methods for allowing a user to dynamically manipulate stereoscopic parameters |
US20080226160A1 (en) | 2007-03-12 | 2008-09-18 | Conversion Works, Inc. | Systems and methods for filling light in frames during 2-d to 3-d image conversion |
US20080227075A1 (en) | 2007-03-15 | 2008-09-18 | Ctb/Mcgraw-Hill, Llc | Method and system for redundant data capture from scanned documents |
US8213711B2 (en) | 2007-04-03 | 2012-07-03 | Her Majesty The Queen In Right Of Canada As Represented By The Minister Of Industry, Through The Communications Research Centre Canada | Method and graphical user interface for modifying depth maps |
US8488868B2 (en) | 2007-04-03 | 2013-07-16 | Her Majesty The Queen In Right Of Canada, As Represented By The Minister Of Industry, Through The Communications Research Centre Canada | Generation of a depth map from a monoscopic color image for rendering stereoscopic still and video images |
US20090002368A1 (en) | 2007-06-26 | 2009-01-01 | Nokia Corporation | Method, apparatus and a computer program product for utilizing a graphical processing unit to provide depth information for autostereoscopic display |
JP4835659B2 (en) | 2007-07-30 | 2011-12-14 | コワングウーン ユニバーシティー リサーチ インスティテュート フォー インダストリー コーオペレーション | 2D-3D combined display method and apparatus with integrated video background |
US7630533B2 (en) | 2007-09-20 | 2009-12-08 | Hologic, Inc. | Breast tomosynthesis with display of highlighted suspected calcifications |
US7852461B2 (en) | 2007-11-15 | 2010-12-14 | Microsoft International Holdings B.V. | Dual mode depth imaging |
US7940961B2 (en) | 2007-12-20 | 2011-05-10 | The United States Of America As Represented By The Secretary Of The Navy | Method for enhancing ground-based detection of a moving object |
US20090219383A1 (en) | 2007-12-21 | 2009-09-03 | Charles Gregory Passmore | Image depth augmentation system and method |
US8265425B2 (en) | 2008-05-20 | 2012-09-11 | Honda Motor Co., Ltd. | Rectangular table detection using hybrid RGB and depth camera sensors |
US8073243B2 (en) | 2008-05-30 | 2011-12-06 | General Instrument Corporation | Replacing image information in a captured image |
EP2327010A2 (en) | 2008-08-22 | 2011-06-01 | Google, Inc. | Navigation in a three dimensional environment on a mobile device |
US8194102B2 (en) | 2008-10-06 | 2012-06-05 | Microsoft Corporation | Rendering annotations for images |
KR101574068B1 (en) | 2008-12-26 | 2015-12-03 | 삼성전자주식회사 | Image processing method and apparatus |
US8325179B2 (en) | 2009-03-04 | 2012-12-04 | Landmark Graphics Corporation | Three-dimensional visualization of images in the earth's subsurface |
US20100259610A1 (en) | 2009-04-08 | 2010-10-14 | Celsia, Llc | Two-Dimensional Display Synced with Real World Object Movement |
WO2010150936A1 (en) | 2009-06-24 | 2010-12-29 | (주)엘지전자 | Stereoscopic image reproduction device and method for providing 3d user interface |
US8320634B2 (en) | 2009-07-11 | 2012-11-27 | Richard Deutsch | System and method for monitoring protective garments |
CA2772607A1 (en) | 2009-09-01 | 2011-03-10 | Prime Focus Vfx Services Ii Inc. | System and process for transforming two-dimensional images into three-dimensional images |
WO2011029209A2 (en) | 2009-09-10 | 2011-03-17 | Liberovision Ag | Method and apparatus for generating and processing depth-enhanced images |
US8659592B2 (en) * | 2009-09-24 | 2014-02-25 | Shenzhen Tcl New Technology Ltd | 2D to 3D video conversion |
US8947422B2 (en) | 2009-09-30 | 2015-02-03 | Disney Enterprises, Inc. | Gradient modeling toolkit for sculpting stereoscopic depth models for converting 2-D images into stereoscopic 3-D images |
US8537200B2 (en) * | 2009-10-23 | 2013-09-17 | Qualcomm Incorporated | Depth map generation techniques for conversion of 2D video data to 3D video data |
US20120306874A1 (en) | 2009-12-14 | 2012-12-06 | Agency For Science, Technology And Research | Method and system for single view image 3 d face synthesis |
JP5387377B2 (en) | 2009-12-14 | 2014-01-15 | ソニー株式会社 | Image processing apparatus, image processing method, and program |
US8472746B2 (en) | 2010-02-04 | 2013-06-25 | Sony Corporation | Fast depth map generation for 2D to 3D conversion |
US20130051659A1 (en) | 2010-04-28 | 2013-02-28 | Panasonic Corporation | Stereoscopic image processing device and stereoscopic image processing method |
KR101198320B1 (en) | 2010-06-21 | 2012-11-06 | (주)아이아이에스티 | Method and apparatus for converting 2d image into 3d image |
EP2416578A3 (en) * | 2010-08-02 | 2014-09-24 | Trdimize Ltd | Multiclass clustering with side information from multiple sources and the application of converting 2D video to 3D |
WO2012016600A1 (en) | 2010-08-06 | 2012-02-09 | Trident Microsystems, Inc. | Method for generating of a depth map, method for converting a two-dimensional image sequence and device for generating a stereoscopic image |
US8428342B2 (en) | 2010-08-12 | 2013-04-23 | At&T Intellectual Property I, L.P. | Apparatus and method for providing three dimensional media content |
WO2012037685A1 (en) | 2010-09-22 | 2012-03-29 | Berfort Management Inc. | Generating 3d stereoscopic content from monoscopic video content |
GB2487039A (en) | 2010-10-11 | 2012-07-11 | Michele Sciolette | Visualizing Illustrated Books And Comics On Digital Devices |
CN102696054B (en) * | 2010-11-10 | 2016-08-03 | 松下知识产权经营株式会社 | Depth information generation device, depth information generating method and stereo-picture converting means |
US9595127B2 (en) | 2010-12-22 | 2017-03-14 | Zspace, Inc. | Three-dimensional collaboration |
KR20120095059A (en) | 2011-02-18 | 2012-08-28 | (주)스튜디오 로프트 | Method of converting 2d images to 3d images |
JP2012190299A (en) | 2011-03-11 | 2012-10-04 | Sony Corp | Image processing system and method, and program |
US8654181B2 (en) | 2011-03-28 | 2014-02-18 | Avid Technology, Inc. | Methods for detecting, visualizing, and correcting the perceived depth of a multicamera image sequence |
JP5891426B2 (en) | 2011-03-31 | 2016-03-23 | パナソニックIpマネジメント株式会社 | An image drawing apparatus, an image drawing method, and an image drawing program for drawing an all-around stereoscopic image |
US20120274626A1 (en) | 2011-04-29 | 2012-11-01 | Himax Media Solutions, Inc. | Stereoscopic Image Generating Apparatus and Method |
US8655055B2 (en) | 2011-05-04 | 2014-02-18 | Texas Instruments Incorporated | Method, system and computer program product for converting a 2D image into a 3D image |
US20120306849A1 (en) | 2011-05-31 | 2012-12-06 | General Electric Company | Method and system for indicating the depth of a 3d cursor in a volume-rendered image |
US20130044192A1 (en) * | 2011-08-17 | 2013-02-21 | Google Inc. | Converting 3d video into 2d video based on identification of format type of 3d video and providing either 2d or 3d video based on identification of display device type |
US8928729B2 (en) * | 2011-09-09 | 2015-01-06 | Disney Enterprises, Inc. | Systems and methods for converting video |
KR101869872B1 (en) | 2011-12-01 | 2018-06-21 | 엘지디스플레이 주식회사 | Method of multi-view image formation and stereoscopic image display device using the same |
US20140307062A1 (en) | 2011-12-05 | 2014-10-16 | Rotary View Ltd. | System and method for generating a stereoscopic 3d presentation from picture sequence emanating from single lens source |
KR101356544B1 (en) * | 2012-03-29 | 2014-02-19 | 한국과학기술원 | Method and apparatus for generating 3d stereoscopic image |
US8644596B1 (en) | 2012-06-19 | 2014-02-04 | Google Inc. | Conversion of monoscopic visual content using image-depth database |
-
2015
- 2015-12-14 US US14/967,939 patent/US9609307B1/en active Active
Cited By (49)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20170109888A1 (en) * | 2015-10-20 | 2017-04-20 | Hewlett-Packard Development Company, L.P. | Alignment of images of a three-dimensional object |
US10447999B2 (en) * | 2015-10-20 | 2019-10-15 | Hewlett-Packard Development Company, L.P. | Alignment of images of a three-dimensional object |
US11574453B2 (en) | 2015-12-15 | 2023-02-07 | Tahoe Research, Ltd. | Generation of synthetic 3-dimensional object images for recognition systems |
US10068385B2 (en) | 2015-12-15 | 2018-09-04 | Intel Corporation | Generation of synthetic 3-dimensional object images for recognition systems |
US10769862B2 (en) | 2015-12-15 | 2020-09-08 | Intel Corporation | Generation of synthetic 3-dimensional object images for recognition systems |
US10366499B2 (en) * | 2016-09-23 | 2019-07-30 | Robotis Co., Ltd. | Device, method, and computer-readable recording medium for editing and playing robot motion |
US20220132222A1 (en) * | 2016-09-27 | 2022-04-28 | Clarifai, Inc. | Prediction model training via live stream concept association |
US11917268B2 (en) * | 2016-09-27 | 2024-02-27 | Clarifai, Inc. | Prediction model training via live stream concept association |
US20180295335A1 (en) * | 2017-04-10 | 2018-10-11 | Red Hen Systems Llc | Stereographic Imaging System Employing A Wide Field, Low Resolution Camera And A Narrow Field, High Resolution Camera |
US20220189130A1 (en) * | 2017-11-29 | 2022-06-16 | Sdc U.S. Smilepay Spv | Systems and methods for constructing a three-dimensional model from two-dimensional images |
US11694418B2 (en) * | 2017-11-29 | 2023-07-04 | Sdc U.S. Smilepay Spv | Systems and methods for constructing a three-dimensional model from two-dimensional images |
US11893755B2 (en) | 2018-01-19 | 2024-02-06 | Interdigital Vc Holdings, Inc. | Multi-focal planes with varying positions |
KR20190130480A (en) * | 2018-04-24 | 2019-11-22 | 주식회사 지디에프랩 | Video resolution enhancement system using change macro-block extraction technique |
WO2019209008A1 (en) * | 2018-04-24 | 2019-10-31 | 주식회사 지디에프랩 | System for improving video quality by using changed macroblock extraction technique |
KR102179436B1 (en) * | 2018-04-24 | 2020-11-16 | 주식회사 지디에프랩 | Video resolution enhancement system using change macro-block extraction technique |
US10657712B2 (en) * | 2018-05-25 | 2020-05-19 | Lowe's Companies, Inc. | System and techniques for automated mesh retopology |
US11689709B2 (en) | 2018-07-05 | 2023-06-27 | Interdigital Vc Holdings, Inc. | Method and system for near-eye focal plane overlays for 3D perception of content on 2D displays |
US20210235067A1 (en) * | 2018-07-06 | 2021-07-29 | Pcms Holdings, Inc. | Method and system for forming extended focal planes for large viewpoint changes |
US11755956B2 (en) * | 2018-08-06 | 2023-09-12 | Samsung Electronics Co., Ltd. | Method, storage medium and apparatus for converting 2D picture set to 3D model |
US20210279971A1 (en) * | 2018-08-06 | 2021-09-09 | Samsung Electronics Co., Ltd. | Method, storage medium and apparatus for converting 2d picture set to 3d model |
WO2020032354A1 (en) * | 2018-08-06 | 2020-02-13 | Samsung Electronics Co., Ltd. | Method, storage medium and apparatus for converting 2d picture set to 3d model |
US11902496B2 (en) * | 2018-12-05 | 2024-02-13 | Fotonation Limited | Depth sensing camera system |
US10991156B2 (en) * | 2018-12-05 | 2021-04-27 | Sri International | Multi-modal data fusion for enhanced 3D perception for platforms |
US20200186782A1 (en) * | 2018-12-05 | 2020-06-11 | Fotonation Ltd. | Depth sensing camera system |
US10839481B1 (en) * | 2018-12-07 | 2020-11-17 | Bellus 3D, Inc. | Automatic marker-less alignment of digital 3D face and jaw models |
US10810738B1 (en) * | 2018-12-07 | 2020-10-20 | Bellus 3D, Inc. | Marker-less alignment of digital 3D face and jaw models |
US11393164B2 (en) | 2019-05-06 | 2022-07-19 | Apple Inc. | Device, method, and graphical user interface for generating CGR objects |
WO2020231640A1 (en) * | 2019-05-13 | 2020-11-19 | Light Field Lab, Inc. | Light field display system for performance events |
CN113825550A (en) * | 2019-05-13 | 2021-12-21 | 光场实验室公司 | Light field display system for performance events |
US11428933B2 (en) | 2019-05-13 | 2022-08-30 | Light Field Lab, Inc. | Light field display system for performance events |
US11783539B2 (en) | 2019-05-17 | 2023-10-10 | SmileDirectClub LLC | Three-dimensional modeling toolkit |
US11470298B2 (en) * | 2019-11-11 | 2022-10-11 | Samsung Electronics Co., Ltd. | Method and apparatus with updating of algorithm for generating disparity image |
US20220028162A1 (en) * | 2019-11-26 | 2022-01-27 | Sdc U.S. Smilepay Spv | Systems and methods for constructing a dental arch image using a machine learning model |
US11900538B2 (en) * | 2019-11-26 | 2024-02-13 | Sdc U.S. Smilepay Spv | Systems and methods for constructing a dental arch image using a machine learning model |
US11850113B2 (en) | 2019-11-26 | 2023-12-26 | Sdc U.S. Smilepay Spv | Systems and methods for constructing a three-dimensional model from two-dimensional images |
US11100658B2 (en) * | 2020-01-10 | 2021-08-24 | United States Of America As Represented By The Secretary Of The Navy | Image georectification with mobile object removal |
US20210352261A1 (en) * | 2020-05-11 | 2021-11-11 | Niantic, Inc. | Generating stereo image date from moncular images |
US11805236B2 (en) * | 2020-05-11 | 2023-10-31 | Niantic, Inc. | Generating stereo image data from monocular images |
CN111935528A (en) * | 2020-06-22 | 2020-11-13 | 北京百度网讯科技有限公司 | Video generation method and device |
US20220051372A1 (en) * | 2020-08-12 | 2022-02-17 | Niantic, Inc. | Feature matching using features extracted from perspective corrected image |
CN112019828A (en) * | 2020-08-14 | 2020-12-01 | 上海网达软件股份有限公司 | Method for converting 2D (two-dimensional) video into 3D video |
US11880990B2 (en) | 2020-09-21 | 2024-01-23 | Samsung Electronics Co., Ltd. | Method and apparatus with feature embedding |
US11871137B2 (en) * | 2020-09-30 | 2024-01-09 | Lemon Inc. | Method and apparatus for converting picture into video, and device and storage medium |
EP4070249A4 (en) * | 2020-12-16 | 2023-03-29 | Tencent America LLC | Reference of neural network model for adaptation of 2d video for streaming to heterogeneous client end-points |
JP2023516305A (en) * | 2020-12-16 | 2023-04-19 | テンセント・アメリカ・エルエルシー | Reference Neural Network Model for Adaptation of 2D Video for Streaming to Heterogeneous Client Endpoints |
JP7447293B2 (en) | 2020-12-16 | 2024-03-11 | テンセント・アメリカ・エルエルシー | References to Neural Network Models for Adaptation of 2D Video for Streaming to Heterogeneous Client Endpoints |
WO2022178293A1 (en) * | 2021-02-19 | 2022-08-25 | Insurance Services Office, Inc. | Computer vision systems and methods for supplying missing point data in point clouds derived from stereoscopic image pairs |
CN113177908A (en) * | 2021-04-01 | 2021-07-27 | 柳城县迪森人造板有限公司 | Identification method and device for solid wood ecological plate |
CN114120149A (en) * | 2021-11-09 | 2022-03-01 | 肇庆市城市规划设计院 | Oblique photogrammetry building feature point extraction method and device, electronic equipment and medium |
Also Published As
Publication number | Publication date |
---|---|
US9609307B1 (en) | 2017-03-28 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US9609307B1 (en) | Method of converting 2D video to 3D video using machine learning | |
US9438878B2 (en) | Method of converting 2D video to 3D video using 3D object models | |
US11721067B2 (en) | System and method for virtual modeling of indoor scenes from imagery | |
US9241147B2 (en) | External depth map transformation method for conversion of two-dimensional images to stereoscopic images | |
Wang et al. | Stereobrush: interactive 2d to 3d conversion using discontinuous warps | |
CN108876814B (en) | Method for generating attitude flow image | |
US11562536B2 (en) | Methods and systems for personalized 3D head model deformation | |
US11315313B2 (en) | Methods, devices and computer program products for generating 3D models | |
KR100560464B1 (en) | Multi-view display system with viewpoint adaptation | |
CN111008927B (en) | Face replacement method, storage medium and terminal equipment | |
EP3474185B1 (en) | Classification of 2d images according to types of 3d arrangement | |
US11587288B2 (en) | Methods and systems for constructing facial position map | |
JP7462120B2 (en) | Method, system and computer program for extracting color from two-dimensional (2D) facial images | |
US20230130281A1 (en) | Figure-Ground Neural Radiance Fields For Three-Dimensional Object Category Modelling | |
Onizuka et al. | Landmark-guided deformation transfer of template facial expressions for automatic generation of avatar blendshapes | |
JP5210416B2 (en) | Stereoscopic image generating apparatus, stereoscopic image generating method, program, and recording medium | |
Patterson et al. | Landmark-based re-topology of stereo-pair acquired face meshes | |
CN115457171A (en) | Efficient expression migration method adopting base expression space transformation | |
Rotman et al. | A depth restoration occlusionless temporal dataset | |
Kang | A structure from motion approach using constrained deformable models and appearance prediction | |
Karpouzis et al. | Compact 3D model generation based on 2D views of human faces: application to face recognition | |
US20220309733A1 (en) | Surface texturing from multiple cameras | |
Finnie | Real-Time Dynamic Full Scene Reconstruction Using a Heterogeneous Sensor System | |
Rochow et al. | FaDIV-Syn: Fast Depth-Independent View Synthesis using Soft Masks and Implicit Blending | |
CN117333914A (en) | Single-image-based face key point extraction method and electronic equipment |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
AS | Assignment |
Owner name: LEGEND3D, INC., CALIFORNIA Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNORS:LOPEZ, ANTHONY;MCFARLAND, JACQUELINE;BALDRIDGE, TONY;SIGNING DATES FROM 20151210 TO 20151211;REEL/FRAME:037284/0823 |
|
AS | Assignment |
Owner name: IP SUCCESSOR FUND 21 L.P., CANADA Free format text: SECURITY INTEREST;ASSIGNOR:LEGEND3D, INC.;REEL/FRAME:040589/0472 Effective date: 20161130 |
|
STCF | Information on status: patent grant |
Free format text: PATENTED CASE |
|
AS | Assignment |
Owner name: IP SUCCESSOR FUND 21 L.P., CANADA Free format text: SECURITY AGREEMENT;ASSIGNOR:PAR INVESTMENT PARTNERS, L.P.;REEL/FRAME:042198/0013 Effective date: 20160818 |
|
FEPP | Fee payment procedure |
Free format text: MAINTENANCE FEE REMINDER MAILED (ORIGINAL EVENT CODE: REM.); ENTITY STATUS OF PATENT OWNER: SMALL ENTITY |
|
FEPP | Fee payment procedure |
Free format text: SURCHARGE FOR LATE PAYMENT, SMALL ENTITY (ORIGINAL EVENT CODE: M2554); ENTITY STATUS OF PATENT OWNER: SMALL ENTITY |
|
MAFP | Maintenance fee payment |
Free format text: PAYMENT OF MAINTENANCE FEE, 4TH YR, SMALL ENTITY (ORIGINAL EVENT CODE: M2551); ENTITY STATUS OF PATENT OWNER: SMALL ENTITY Year of fee payment: 4 |
|
AS | Assignment |
Owner name: LEGEND3D, INC., CALIFORNIA Free format text: RELEASE BY SECURED PARTY;ASSIGNOR:IP SUCCESSOR FUND 21 L.P.;REEL/FRAME:056385/0697 Effective date: 20210510 Owner name: LEGEND3D, INC., CALIFORNIA Free format text: RELEASE BY SECURED PARTY;ASSIGNOR:IP SUCCESSOR FUND 21 L.P.;REEL/FRAME:056422/0735 Effective date: 20210510 |
|
AS | Assignment |
Owner name: USFT PATENTS, INC., DELAWARE Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNOR:LEGEND3D, INC.;REEL/FRAME:063617/0664 Effective date: 20210512 |