WO2001045103A1

WO2001045103A1 - Method and apparatus to identify content stored on a storage medium

Info

Publication number: WO2001045103A1
Application number: PCT/EP2000/011923
Authority: WO
Inventors: Evgeniy Leyvi; Nevenka Dimitrova
Original assignee: Koninklijke Philips Electronics N.V.
Priority date: 1999-12-14
Filing date: 2000-11-29
Publication date: 2001-06-21

Abstract

A system that utilizes a determined unique characteristic from content for subsequently identifying the content and/or the corresponding content review position in the time domain. The system utilizes a database of previously stored unique characteristics from known content for matching to the determined unique characteristic from the content source. The unique characteristics may correspond to video tape indexing data, keyframe data, audio characteristics, text occurrences, and/or other known characteristics from the content. When a match is found between the determined unique characteristic and the stored unique characteristics, the system identifies the content. When corresponding time domain data is stored in the database, the system identifies the current content review position in the time domain. When the unique characteristic from the content does not match any of the unique characteristics stored in the database, the system stores the current unique characteristic and corresponding content identifying information in the database for identifying the content at some subsequent time.

Description

Method and apparatus to identify content stored on a storage medium

Field of the Invention

This invention generally relates to a method and apparatus for identifying content stored on a storage medium. Specifically, the present invention pertains to utilizing stored unique characteristics to identify content and/or a given storage medium on which the content is stored.

Background of the Invention

A problem exists in the usage of storage mediums for storing content that is reviewed in a sequential manner such as time sequential content. With storage mediums, such as a digital video recording device, the storage medium may contain numerous individual content portions that can only be identified by reviewing each individual content portion. In other storage mediums, such as analog vertical-helical scanning (NHS) tapes, generally video tapes, the video tapes may only contain a relatively small number of individual content portions. However, a given user typically utilizes multiple video tapes and therefore the problem is further compounded in that not only must a particular content portion be identified, but a particular video tape on which the content portion is stored must also be identified.

In the prior art, there is no system for automatically identifying a given content portion or a given storage medium. For commercially recorded mediums, the manufacturer typically affixes a label to the storage medium. The label may contain identifying information such as the storage medium content, total content runtime, content production information, etc. However, over time these labels may become separated from the storage medium, such as the video tape cassette, thereby requiring a user to play the video tape and thereby visually identify the content. For user recorded content, the user is required to affix a blank label to the storage medium and note content identifying information on the blank label. Additionally, a user may note content identifying information on a storage medium carton for the purposes of identifying the content on the storage medium. This is a process that oftentimes is neglected by the user and results in storage medium or multiple storage mediums that contain recorded content that is not identifiable without visually reviewing the content on a reviewing device, such as a television, audio tape player, computer monitor, etc.

Even in a case wherein the storage medium or the storage medium carton contains information identifying the content of the storage medium, there is no ready way to identify a current viewing position of the content without starting the content at a beginning position. In the case of a digital storage medium, some positional information, such as frame data, etc., may be available for prerecorded content, but typically is not available for user recorded content. After starting the content portion from the beginning, such as by rewinding a video tape, a counter may be reset to identify the time sequential beginning of the storage medium. For example, for video tape, the video tape may be rewound to the beginning.

Regardless of the type of storage medium, resetting the counter at the content initial position causes the counter to contain storage medium time positional information that corresponds to the time sequential beginning of the content portion and/or the storage medium. Thereafter, when the content portion is played in a time sequential manner, fast- forwarded, etc., the counter will contain content time positional information that will correspond to the current time position of the content. However, if the current storage medium is replaced with a new storage medium, the above procedure must be repeated for the new storage medium so that the counter will contain the proper time positional information for the content contained on the new storage medium. Time positional information is particularly important in a case wherein the storage medium contains more than one recorded program. In this case, when the user wishes to review content that does not start at a known place on the storage medium, the user may ensure that the storage medium begins at the initial position before the user resets the counter to reflect the storage medium initial position. For example, the user may rewind the video tape, etc., before resetting the video tape counter. Alternatively, the user may review a content portion at a current position, make a rough determination of the current sequential position of the reviewed content, and then reposition the sequential position of the content to find a desired content position. As a third alternative, the user may review the content at a normal or accelerated reviewing speed while sequentially repositioning the content to find the desired content position.

All of the above procedures are particularly cumbersome for a user. In some consumer equipment, some positional information may be available from the storage medium. For example, in some video cassette recorders (NCRs), indexing information can be recorded directly on a control track of a video tape. In the Video Indexing Search System (NISS), an indexing mark is placed on the control track of the video tape each time recording begins. The recorded index marks are not distinguishable from each other. In this way, if a user wishes to view the third recorded program on the video tape, the user can enter a fast forward or rewind mode to go a particular number of index marks forward or backward. However, this system has the disadvantage in that the user must still first identify where the video tape is presently positioned to determine how many index marks there are between the current position and the desired position on the video tape. In addition, this system only enables a user to proceed to the beginning portion of a content portion or to a content position wherein the user specified an additional index mark at the time of recording. This system does not enable a user to identify the content of the video tape or the current video tape position in the time domain.

In an alternate known system, Video Address Search System (VASS) pulses are stored on the control track of the video tape. The VASS pulses have an advantage over the VISS pulses in that the VASS pulses are distinct from each other. Accordingly, a user can fast forward or rewind to a given VASS pulse (e.g., go to VASS pulse three) without having to determine where the video tape currently resides. However, this system still does not help a user identify the video tape or the current video tape position in the time domain. In fact, both the VISS and the VASS pulses are more akin to a bookmarking system that provides no information on the recorded content. In yet another known system, a time code is recorded on the control track of the video tape. This time code can be used for identifying the current position of the video tape in the time domain but does not help a user identify content stored on the video tape. For a digital storage medium, time domain positional information may be available from the storage medium yet again this does not help a user identify content stored on the storage medium.

In other known systems, keyframes are extracted from video sequences to facilitate video editing, storage, retrieval, commercial skipping, etc. For example, in U.S. Patent Application Νos. 08/867,140, 08/867,145, 09/220,277, and 09/123,444, to Dimitrova et. al., incorporated herein by reference, keyframes are identified through the use of techniques such as cut detection, static scene detection, similar frame detection, black frame detection, text detection, commercial detection, etc. Extracted keyframes from a video sequence may help a user to identify the corresponding video sequence but do not help automatically identify the content or the storage mediums current reviewing position in the time domain. Accordingly, it is an object of the present invention to overcome the disadvantages of the prior art.

It is a further object of the present invention to provide a method of facilitating the use and operation of a storage medium.

Summary of the Invention

These and other objects of the present invention are achieved by a system that utilizes determined unique characteristics from content for subsequently identifying the content and/or the corresponding content review position in the time domain. The system utilizes a database of previously stored unique characteristics from known content for matching to a determined unique characteristic(s) from currently reviewed content. The unique characteristic(s) may correspond to video tape indexing data, keyframe data, audio characteristics, text occurrences, and or other known unique characteristic(s) from the content. When a match is found between the determined unique characteristic(s) and the stored unique characteristics, the system identifies the content. In addition, if corresponding time domain data is stored in the database, the system identifies the current content review position in the time domain.

In accordance with the present invention, when the determined unique characteristic(s) from the content do not match any of the stored unique characteristics in the database, the system stores the current unique characteristic(s) and corresponding content identifying information in the database for identifying the content at some subsequent time.

The unique characteristic(s) from the content may be all one type of unique characteristic, such as index marks, keyframes, text occurrences, scene transitions, a given audio characteristic, etc. In addition, the unique characteristic(s) may be any combination of the above and other known unique characteristic types.

Brief Description of the Drawings

The following are descriptions of embodiments of the present invention that when taken in conjunction with the following drawings will demonstrate the above noted features and advantages, as well as further ones. It should be expressly understood that the drawings are included for illustrative purposes and do not represent the scope of the present invention. The invention is best understood in conjunction with the accompanying drawing in which: FIG. 1 shows an illustrative identifying system in accordance with an embodiment of the present invention.

Detailed Description of the Invention In the discussion to follow, certain terms will be illustratively discussed in regard to specific embodiments or systems to facilitate the discussion. However, as would be readily apparent to a person of ordinary in the art, these terms should be understood to encompass other similar applications and embodiments wherein the present invention could be readily applied. For example, the term content should be understood to include video content, audio content, text content, animation content, or any other content that may be reviewed, received, recorded, played, etc., in a given sequential manner. In addition, content should be understood to include any combination of the above content types, such as audio visual content. The terms storage medium, memory, etc., should be understood to include linear storage mediums, such as video tapes, audio tapes, etc., random access storage . tediums, such as hard disks, digital video discs, other optical storage mediums, integrated circuit storage mediums, etc. The term sequential content should be understood to mean that the content, in some mode of operation, may be reviewed (e.g., watched, listened to, read, etc.) in a sequential manner in the time domain. For example, sequential content should be understood to include all of the above time sequential content. In addition, sequential content may not strictly depict content occurring in a time sequential manner, such as time lapse content, content that changes time references, etc. A time sequential beginning of content should be understood to include setting a storage medium to be reviewed at an initial position of a stored content portion. A unique content characteristic should be understood to include any unique characteristic that may be detected from a content portion. For instance, scene transitions in video content may be detected, black frames may be detected, the occurrence of text may be detected, the occurrence, position, size, and motion of distinct image regions (e.g., face regions, text regions, etc.) may be detected. In addition, it is known in the art that these unique characteristics of the video content may be utilized for detecting particular frames of the content such as keyframes. For audio content, audio signal characteristics such as short time energy, band energy ratio, pause rate, pitch, signal amplitude, etc. may be detected during content portions. For the above and for other content, other known unique characteristics may also be detected.

FIG. 1 depicts an illustrative identification system 100 in accordance with an embodiment of the present invention. For the purposes of the following discussion, the term play will be utilized herein below as an example of a reviewing process to facilitate the discussion to follow. However, the term play should be understood to encompass all manners of reviewing content portions in a sequential manner. In accordance with the present invention, the identification system 100 retrieves a unique characteristic of a particular frame or frames from a content portion to enable identification of a storage medium on which the content portion is stored, the content portion stored thereon, and/or the content portions current play position in the time domain.

The identification system 100 contains a processor 120 having a content characteristic capture device, such as a frame capture 130. The processor 120 utilizes the frame capture 130 for buffering and storing data corresponding to a frame or frames of a video content portion. The processor 120, utilizing the frame(s) stored in the frame capture 130 detects unique characteristics of the video content portion. It is known in the art that many different unique characteristics may be detected from a video content portion including video frame data, partial video frame data, video frame region data, motion data, frame signature data, etc., (as further discussed above). In addition, the occurrence of data in a control section of a storage medium may be utilized for selecting a given content portion. In this embodiment, the data in the control section may illustratively be an index mark. When the index mark is detected by the processor 120, the processor may select a frame corresponding to the index mark and thereafter treat the selected frame as a unique characteristic in accordance with the present invention.

In the following discussion, keyframes are illustratively discussed as one known type of unique characteristic that may be detected from video content. However, it should be understood that keyframes are merely discussed herein below in terms of the present invention for the purposes of facilitating the discussion. As would be readily apparent to a person of ordinary skill in the art, any system in which unique characteristics that are associated with (e.g., contained within), part of, and/or may be determined from a given content portion and content type, may also by utilized in accordance with the present invention. Accordingly, in the following discussion, the keyframe should be understood to encompass these other unique characteristics of content that are known in the art. The identification system 100 operates utilizing the premise that content portions contain unique characteristics that may be determined from the content portions. Since each of these determined characteristics are unique, any one or more of the unique characteristics may be utilized for identifying a corresponding medium, such as a video cassette/tape and the content stored therein/thereon. The identification system 100 also contains a memory 140 and a time reference 150. The memory 140 is utilized for storing unique characteristic information determined from the content portion. The memory 140 may also store time domain information corresponding to the unique characteristic's play position in the time domain. In addition, the memory 140 may store content identification information of corresponding content portions as described in more detail below.

As an example, the data stored in the memory 140 may be detected from previously played content portions. Previously played content is just one type of known content. Herein, the term known content should be understood to include content wherein the identification system has previously stored unique characteristic data corresponding to the content regardless of the source of the unique characteristic data. For instance, the unique characteristics and corresponding content identifying information may be provided to the memory 140 from an alternate input 142, such as an Internet connection, for prerecorded content. In an alternate embodiment, the unique characteristics and corresponding content identifying information may be included at an initial position of a content portion, etc.

In operation, a medium containing content (e.g., a video cassette) is placed in a content source, such as a video cassette recorder (VCR) 110, and a play function is initiated. The play function may be initiated, for instance, by depressing a play button on the VCR 110, by depressing a play button on a remote control (not shown), by initiating a cassette tape query or play function on the identification system 100, or any other known system. After a play function is initiated, the processor 120 receives content data from the video tape through a data line 160. The content data contains unique characteristics, such as identified keyframes, that may be detected by the processor 120 for the purposes of operating in accordance with the present invention. The processor 120, in accordance with the present invention, utilizes the keyframes to automatically identify the video content, the video cassette tape position that corresponds to each keyframe, and/or the tape identity as described in more detail below. The processor 120 may select/identify a keyframe using any of the many ways in which it is known to select keyframes in the art including known video cut detection processes. Whichever known system is utilized by the processor 120 for selecting the keyframes from the video content should be utilized each time so that the identifying system 100 can repeatedly select the same keyframes from a given content portion.

After the frame capture receives a frame that the processor 120 thereafter identifies as a keyframe, illustratively the processor 120 may derive a signature of the DCT coefficients of the keyframe. During the keyframe selection process, these frame signatures may be utilized to represent the keyframe and/or to analyze the frame content. The signatures may also be utilized for spotting patterns within the keyframe that may also be utilized as unique characteristics of the video content. For example, keyframe signature may be derived for each grouping of similarly valued DCT blocks in a frame. A keyframe signature may be derived from region signatures that in turn are derived from block signatures within the keyframe. Qualitatively, the frame signatures may be examined to determine unique characteristics about prominent regions that represent identifiable objects in the conesponding frames. These objects may include superimposed text boxes (e.g., in news broadcasts, credits, etc.), people's faces, etc. In these embodiments, the frames may be examined based on the size of the prominent objects. The frame comparison may be based on the strict position of the prominent objects to eliminate all or all but one of a visually similar set of keyframes. In other embodiments, the frames may be examined without regard to the arrangement of prominent objects. Once a unique characteristic, such as a keyframe is identified, the processor

120 may illustratively determine a signature for the keyframe and thereafter utilize the keyframe signature as the determined unique characteristic. The processor 120 then queries the memory 140 to determine if the currently identified keyframe signature conesponds to any previous keyframe signatures stored in the memory 140. In the event that the currently identified keyframe signature does not conespond to a previously stored keyframe signature, then the processor 120 treats the content portion as an unknown content portion and the identification system 100 stores the cunently identified keyframe signature. In addition, the identification system 100 may associate tape identifying information together with the currently identified keyframe signature, for instance, by querying the user to provide the tape identifying information. The tape identifying information may include a title for the video cassette and also may include a description of any or all of the content stored thereon. The identification system 100 then stores the cunently identified keyframe signature and the tape identifying information together in the memory 140, in associated portions of the memory 140, or in a separate associated memory (not shown).

In a particular embodiment, when the currently identified keyframe signature does not conespond to any keyframe signatures previously stored in the memory 140, the identification system 100 may command the VCR 110 to scan through more of the video tape. In this way, the processor 120 may identify additional keyframe signatures that are on the video tape and store, in the memory 140, the additional keyframe signatures together with the corresponding tape identifying information.

During the time between identification of keyframe signatures, the identification system 100 may interact with the VCR 110, for instance through use of a transmitter/receiver (e.g., infra-red transmitter/receiver) 125, 115 or directly through a wired control connection (not shown). As an example, the identification system 100 may send a command to the VCR 110 to put the VCR 110 into an fast forward scanning mode, an accelerated viewing speed, or any other accelerated video tape playing mode to reduce the time between keyframe signatures. Any of the above modes may be utilized as long as the VCR 110 can identify keyframe signatures from the video cassette during the accelerated video tape playing mode. The identification system 100 may command the VCR 110 to go into the mode that is best supported by the VCR 110 (e.g., which ever mode results in the shortest time between identified keyframe signatures). The identification system 100 may be instructed as to which mode is supported by the VCR 110, for instance during an initial setup of the identification system 100.

In yet another embodiment, the processor 120 may store each keyframe signature determined from the video tape. In this way, the identification system 100 may identify a video tape in the least amount of time regardless of the video tapes position at the time that it is inserted into the VCR 110. In this embodiment, the tape identification system 100 may cause the video cassette to rewind to the beginning or some other determined portion of the video tape. At this time, the time reference 150, or a time reference or displayed index reference resident on the VCR 110 may be initialized. In another embodiment, the processor 120 may note the cunent indication of the time reference 150, etc., instead of resetting the time reference 150. Thereafter, as the identification system 100 identifies each keyframe on the video tape and determines the conesponding keyframe signature, the processor 120 may also store the conesponding time or a conesponding indication of the video tape in the time domain at the occurrence of the keyframe signature.

In the event that the cunently identified keyframe signature does conespond to a previously stored keyframe signature, then the video portion is a known video portion. For a known video portion, the processor 120 may identify the cunent video portion, the video tape, and/or may provide corresponding tape identifying information to the user. In the event that the identification system 100 has previously stored the conesponding time or the indication of the video tape in the time domain at the occurrence of the identified keyframe signature, the identification system 100 may also identify the cunent video tape position in the time domain. Alternatively, the identification system 100 may only identify the video tape position in the time domain without providing corresponding tape identifying information to the user.

In one embodiment, the identification system 100 displays the cunent video tape identifying information and/or video tape position on a display 170. The display 170 may be a digital display such as a light emitting diode (LED) display, a liquid crystal display

(LCD), a cathode ray tube (CRT) display, such as a television display, etc., and may be incorporated into or separate from the identification system 100. In an other embodiment, the identification system 100 controls the VCR 110 to reset video tape positional indicators resident on the VCR 110 to indicate the proper video tape positional information in the time domain. In an other embodiment, the identification system 100 controls the VCR 110 to indicate the proper video tape positional information as an absolute number not related to time. As previously discussed, the identification system 100 may control the VCR 110 through the use of the transmitter/receiver 125, 115 or by sending control information to the VCR 110 through the data line 160 or through the alternate control line (not shown).

In a particular embodiment, in accordance with the present invention, it may be desirable to eliminate some unique characteristics (e.g., selected keyframes) from consideration. For instance, some umque characteristics of the recorded content may be repetitive among different content portions and therefore may lead to improper identification of the content. For example, within recorded video content, portions of the video content may contain commercials that may also be repeated during other portions of the video content or during portions of completely different video content. If unique characteristics of the commercial were extracted from the commercial and where then utilized to identify the video content, a system in accordance with the present invention may enoneously identify one video portion as another. Accordingly, it may be desirable to eliminate certain content portions that may tend to provide enoneous identification results. For example, it is known in the art that particular frames of recorded video content may be first identified and thereafter, the selected frames may be filtered to remove frame data from content sequences that may commonly be found in the recorded content, such as commercial sequences. Some known VCR's record a time code directly on the control track of the video tape. When the identification system 100 is used in conjunction with a VCR that records a time code directly on the control track of the video tape, the identification system 100 may store the time code together with conesponding keyframe signature in the memory 140. In this embodiment, the identification system 100 may utilize the stored time code and keyframe signature for the purposes of matching to subsequent time code and keyframe signatures and thereby, identify the video content at some future time.

Finally, the above-discussion is intended to be merely illustrative of the invention. Those having ordinary skill in the art may devise numerous alternative embodiments that are within the scope of the present invention. For example, although the above discussion illustratively discussed determining unique characteristics identified from video content, the present invention clearly may be suitably utilized for any other content that is known in the art that contains identifiable unique characteristics. For instance, the present invention may be suitably utilized for identifying audio content. Unique characteristics of the audio content may be identified including short time energy, band energy ratio, pause rate, pitch, etc. Other unique characteristics may be identified using Fourier transform and or Mel Spectrum Frequency coefficients. In addition, a weighted combination of these unique characteristics may also be utilized to determine other suitable unique characteristics from the audio content. The unique characteristics may be identified utilizing a sliding window of N milliseconds with an overlapping of M milliseconds. Illustratively N may be selected to be 45 milliseconds and M could be selected as 30 milliseconds. In this embodiment, the characteristic capture device 130 may capture content over a period of time for the purposes of identifying unique characteristics of given content. Regardless of the parameters that are utilized for identifying the unique audio characteristics, as discussed above with regards to unique video characteristics, the same parameters should be utilized each time so that the same unique characteristics are identified each time the audio content is reviewed.

Similarly, multiple keyframe signatures may be first determined from a content portion and thereafter be utilized by the identification system 100 for ensuring the accuracy of the identification of the content portion. In these embodiments, the identification system 100 may query the memory 140 to determine that these multiple keyframe signatures are associated with a known content portion prior to identifying the content portion to a user. In a case wherein the multiple keyframe signatures do not conespond to multiple keyframe signatures from known content, then the identification system 100 will treat the content portion as described above with regard to an unknown content portion. In addition, although in the above discussion, a single type of unique characteristic is identified in a given embodiment (e.g., keyframes, etc.), as would be readily appreciated by a person of ordinary in the art, a multitude of unique characteristics may be identified from a given content portion/type. In an alternate embodiment, these varied characteristics may be utilized in accordance with the above described invention. For example, a keyframe signature may be utilized as one unique characteristic that is stored in the memory 140 and a unique audio characteristic may be stored in the memory 140 as another unique characteristic. In addition, although the source of content is illustratively shown as the VCR 110, clearly other sources of content could readily be utilized in accordance with the present invention. Other sources could include random access memory (RAM), a digital video cassette recorder (DVCR), a non-removable media recorder, such as a TΓVO™, digital versatile discs (DVD) including DVD RAM, compact disc read only/read and write memory (CD-ROM, CDR, CDRW), digital audio tape (DAT), the Internet, cable, satellite, as well as other memory devices as defined herein above. In an embodiment wherein the source of content is a transmission medium

(e.g., cable, Internet connection, etc.), the input 142 may serve as an alternate or additional connection to the transmission medium for receiving content by the processor, for example as shown in FIG. 1. In this embodiment, the content may be stored at a remote site including a remote server, such as an Internet server, a remote service provider site, such as a video on- demand service provider, etc. In these or alternate embodiments, the content may be received from live content feeds, such as a live camera feed, live audio feed, etc. In these embodiments, the memory or another memory may be utilized for storing the content contained within the live content feed.

In the discussion above, the present invention is discussed in terms of identifying content for a user. This should be generally understood to include use of the invention to facilitate identification of content for the purposes of determining if the intellectual property rights of an owner of a given content portion are infringed. As an example, an identifying device in accordance with the present invention may be utilized to automatically identify illegitimate copies of copy protected content contained at an Internet site. In accordance with this embodiment, the memory may contain a program search routine utilized by the processor for searching the Internet via the input 142. Utilizing the program, the processor may retrieve content portions from the Internet and extract unique characteristics as discussed above with regard to other embodiments. In a case wherein a retrieved content portion is identified, the processor may thereafter determine if the source of the content portion is an illegitimate source. For illegitimate sources that are identified by the processor, the processor may produce an indication on the display of the content portion and the source of the content portion.

Although the source of content and the identifying system are illustratively shown as two separate devices, clearly it would be within the capability of a person of ordinary skill in the art to combine these devices into a single device. Further, although the time reference is shown as a separate device from the processor, clearly the time reference may be a function incorporated into the processor as a software, and/or hardware function.

Other combinations of characteristics may also be suitably utilized without departing from the spirit and scope of the following claims.

In interpreting the appended claims, it should be understood that: a) the word "comprising" does not exclude the presence of other elements or acts than those listed in a given claim; b) the word "a" or "an" preceding an element does not exclude the presence of a plurality of such elements; c) any reference signs in the claims do not limit their scope; and d) several "means" may be represented by the same item or hardware or software implemented structure or function.

Claims

CLAIMS:

1. A method of identifying content comprising: determining a unique characteristic from said content, identifying said content using said determined umque characteristic.

2. The method of Claim 1, wherein said identifying said content comprises querying a memory (140) to determine if said determined unique characteristic corresponds to a previously stored unique characteristic.

3. The method of Claim 2, wherein if said unique characteristic does not conespond to said previously stored unique characteristic, said method comprises: storing said determined unique characteristic; and associating tape identifying information to said stored determined unique characteristic.

4. The method of Claim 1, wherein said content is video content and said unique characteristic is one of a keyframe, a keyframe signature, and an identifiable object.

5. The method of Claim 1, wherein said identifying said content comprises identifying said content in the time domain.

6. The method of Claim 1, comprising storing unique characteristics from a plurality of known content sources, wherein said identifying comprises determining if said unique characteristic determined from said content matches any of said stored unique characteristics from said plurality of known content sources.

7. The method of Claim 6, wherein said unique characteristic determined from said content is a first unique characteristic determined from said content, and wherein said storing unique characteristics from said plurality of known content sources comprises storing a plurality of unique characteristics for each of said plurality of known content sources, said method comprising determimng a second unique characteristic from said content, wherein said identifying comprises determining if both said first and second unique characteristics matches two of said plurality of unique characteristics from a given one of said plurality of known content sources.

8. The method of Claim 1, wherein if said unique characteristic corresponds to a filtered portion of said content, said method comprises selecting an alternate unique characteristic from an unfiltered portion of said content, and said identifying comprises identifying said content using said alternate unique characteristic.

9. The method of Claim 7, wherein said first and second unique characteristics are at least one of audio characteristics, text characteristics, and video characteristics.

10. The method of Claim 1, wherein said content is stored on one of an analog storage medium, a local digital storage medium, and a server based storage medium.

11. A method of identifying content comprising : determimng a unique characteristic from said content; comparing said unique characteristic to a database of stored unique characteristics conesponding to a plurality of known content; and identifying said content if said determined unique characteristic matches one of said stored unique characteristics.

12. The method of Claim 11 , wherein if said if said determined unique characteristic does not match one of said stored unique characteristics, said method comprises: storing said unique characteristic from said content; and associating content identifying information to said stored unique characteristic from said content.

13. The method of Claim 12, wherein said associating content identifying information comprises requesting content identifying information from a user.

14. An apparatus for identifying content comprising a processor (120) configured to identify a unique characteristic from said content and configured to identify said content using said identified unique characteristic.

15. The apparatus of Claim 14, comprising a memory configured to store unique characteristics of known content, wherein said processor (120) is configured to query said memory (140) and to identify said content if said identified unique characteristic conesponds to a stored unique characteristic of said known content.

16. The apparatus of Claim 15, wherein said processor (120) is configured such that if said unique characteristic does not conespond to said previously stored unique characteristics of said known content, said processor is configured to: store said identified unique characteristic in said memory (140) ; and associate content identifying information to said stored identified unique characteristic.

17. The apparatus of Claim 14, wherein said content is video content and said unique characteristic is one of a keyframe, a keyframe signatures, and identifiable object.

18. The apparatus of Claim 14, wherein said unique characteristic is one of an audio characteristic, a text occurrence, and a scene transition.

19. The method of Claim 14, wherein said content is stored on one of an analog storage medium, a local digital storage medium, and a server based storage medium.

20. The apparatus of Claim 14, wherein said processor is configured to identify said content in a time domain.