Data processing: database and file management or data structures – Database design – Data structure types
Reexamination Certificate
1998-12-03
2001-12-18
Amsbury, Wayne (Department: 2672)
Data processing: database and file management or data structures
Database design
Data structure types
C707S793000, C707S793000
Reexamination Certificate
active
06332144
ABSTRACT:
FIELD OF THE INVENTION
The present invention relates generally to the field of multimedia and, more particularly, to a technique for annotating media.
BACKGROUND OF THE INVENTION
There are a large number of organizations that presently have substantial amounts of audio, video, and image content in analog form. Many of these organizations are currently moving toward putting such multimedia content into digital form in order to save costs in the areas of data storage and retrieval. That is, similar to other types of data, multimedia content can be easily stored on and retrieved from relatively inexpensive digital storage devices.
The migration of multimedia content from analog form to digital form also provides an organization with the ability to store, search, browse, and retrieve digitized multimedia content from distributed sites. That is, an organization having a number of distributed offices can store, search, browse, and retrieve digitized multimedia content from a centralized storage facility over a proprietary intranet computer network such as, for example, a local area network (LAN), or a public internet computer network such as, for example, the world wide web.
Furthermore, the multimedia content itself may be distributed. That is, an organization that is global in nature may have a number of distributed permanent archival storage locations where digitized multimedia content is permanently stored, or a number of distributed temporary storage locations where digitized multimedia content that is associated with work in progress is temporarily stored. Similar to above, such an organization could also store, search, browse, and retrieve digitized multimedia content from the distributed storage locations over a proprietary intranet computer network or a public internet computer network.
Additionally, an organization may want other entities located outside of the organization to be able to search, browse, and retrieve digitized multimedia content stored and maintained within the organization. For example, an organization may want to sell multimedia content to an outside entity, which may then use the purchased multimedia content for some purpose such as, for example, a news broadcast. Similar to above, the outside entity could search, browse, and retrieve digitized multimedia content from a storage facility within the organization over a proprietary intranet computer network or a public internet computer network.
However, despite the above-described benefits associated with digitized multimedia content, organizations presently have little or no means of searching within multimedia content, organizing information about multimedia content, and delivering multimedia content in a ubiquitous manner. That is, there are presently little or no means for searching inside streams of multimedia content (e.g., audio/video streams), adding meta-information to multimedia content (i.e., annotating multimedia content) for purposes of indexing within multimedia content, and providing universal access to indexed multimedia content over a variety of connection speeds and on a variety of client platforms. Accordingly, it would be desirable to provide a technique for organizing distributed multimedia content and for searching, browsing, and retrieving such organized distributed multimedia content in an efficient and cost-effective manner so as to overcome the above-described shortcomings of the prior art.
OBJECTS OF THE INVENTION
The primary object of the present invention is to provide a technique for annotating media.
The above-stated primary object, as well as other objects, features, and advantages, of the present invention
According to the present invention, an item of audio or video media is annotated by identifying the particular time or times within the period defined by the start and end times of a media stream forming an item of media. The time(s) identified correspond to the time or times at which content within the media stream corresponds to an annotation value, such as a textual item, e.g. a word, or another type of item, e.g. a person or thing. The annotation value is associated with the identified time(s) to annotate the media.
Preferably, each identified time defines the start time of an independently viewable portion of the media stream. Other particular times within the period at which content within the media stream corresponds to a different annotation value may also be identified if desired. This other annotation value can then be associated with the other particular times to annotate the media. The annotation values, although different, may be of the same type or of different types. The independently viewable portions of the media stream having the start time defined by the identified times associated with the respective annotation values will typically overlaps, at least to some extent.
According to another aspect of the invention, a probability representing a measure of confidence in an accuracy of the annotation value corresponding to the identified time(s) is identified. It should be noted that the time(s) within the period at which content within the media stream corresponds to a specific annotation value may be identified more than once. For example, the times might first be determined using an automatic process, such as a daemon process, and then a manual process, such as by a human annotator. Hence, different probabilities could be associated with the same or different identified times for a given annotation value. Advantageously, identified times which were previously associated with a specific annotation value, but have a lower probability then subsequently identified times associated with the same annotation value, are disassociated with the annotation value.
Typically, a processor, in accordance with programmed instructions stored on an electronic, magnetic, optical or other type media, identifies the appropriate times and associates these times with the annotation value. The processor also computes the probabilities or receives the probability as input from the human annotator. The associated annotation values, identified times and probabilities are typically stored in a memory as an annotation index. Multiple distributed processors can be used and each of the processor and the memory storing the annotation index may be located at different sites on a network, such as the Internet.
In a still further aspect of the present invention, the processing device combines the address identifier and the location identifier to generate a combined identifier for use in accessing the item of interest at the location of interest. The combined identifier is beneficially in an extended URL format.
REFERENCES:
patent: 4616336 (1986-10-01), Robertson et al.
patent: 4864501 (1989-09-01), Kucera et al.
patent: 5008853 (1991-04-01), Bly et al.
patent: 5146552 (1992-09-01), Cassorla et al.
patent: 5220657 (1993-06-01), Bly et al.
patent: 5502727 (1996-03-01), Catanzaro et al.
patent: 5581682 (1996-12-01), Anderson et al.
patent: 5600775 (1997-02-01), King et al.
patent: 5625833 (1997-04-01), Levine et al.
patent: 5721827 (1998-02-01), Logan et al.
patent: 5788814 (1998-07-01), Moran et al.
patent: 5822539 (1998-10-01), Van Hoff
patent: 5826025 (1998-10-01), Gramlich
patent: 5920694 (1999-07-01), Carleton et al.
patent: 5938724 (1999-08-01), Pommier et al.
patent: 6006241 (1999-12-01), Purnaveja et al.
Vides and Audio: Organization and Retrieval on the WWWhttp://bspaa.com/vosaic/corp/papers/www5.html.
An Annotation Engine for Supporting Video Database Population http://hulk.bu.edu/pubs/papers/1996/carrer-vane96/TR-08-15-96.html.
Media Stream: An Iconic Visual Language for Video Representation http://web.interval.com/papers/mediastreams/.
deVries Arjen P.
Eberman Brian
Kovalcin David E.
Sokolov Michael
AltaVista Company
Amsbury Wayne
Havan Thu-Thao
Hunton & Williams
LandOfFree
Technique for annotating media does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Technique for annotating media, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Technique for annotating media will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2562268