System and method for creation of video annotations
First Claim
Patent Images
1. A method for annotating a video data stream, at least one portion of the video data stream having associated geographic location data, the method comprising:
- creating a video document by integrating geographic location data into the video data stream, wherein the video document comprises a plurality of video sequences and wherein each of the plurality of video sequences has associated geographic location data;
organizing the video document into a title-and-chapter navigation structure, the navigation structure comprising at least one title and each title comprising at least one chapter, wherein each chapter is associated with at least one of the plurality of video sequences;
extracting a plurality of keyframes from the video data stream each keyframe corresponding to one of the titles and chapters and having associated geographic location data;
mapping a site name for each title and chapter from the geographic location data, wherein the site names for each title have a first level of abstraction, and wherein the site names for each chapter have a second level of abstraction different from the first level of abstraction; and
automatically assigning the site name as a text label to each keyframe of each title and chapter.
1 Assignment
0 Petitions
Accused Products
Abstract
A system and method for annotating a video data stream integrates geographic location data into the video data stream. A site name is mapped from the geographic location data, and the site name is automatically assigned as a text label to the associated portion of the video data stream.
-
Citations
21 Claims
-
1. A method for annotating a video data stream, at least one portion of the video data stream having associated geographic location data, the method comprising:
-
creating a video document by integrating geographic location data into the video data stream, wherein the video document comprises a plurality of video sequences and wherein each of the plurality of video sequences has associated geographic location data; organizing the video document into a title-and-chapter navigation structure, the navigation structure comprising at least one title and each title comprising at least one chapter, wherein each chapter is associated with at least one of the plurality of video sequences; extracting a plurality of keyframes from the video data stream each keyframe corresponding to one of the titles and chapters and having associated geographic location data; mapping a site name for each title and chapter from the geographic location data, wherein the site names for each title have a first level of abstraction, and wherein the site names for each chapter have a second level of abstraction different from the first level of abstraction; and automatically assigning the site name as a text label to each keyframe of each title and chapter. - View Dependent Claims (2)
-
-
3. A method for creating a text label for a video sequence comprising:
-
integrating geographic location data into a video data stream; organizing the video data stream into a title-and-chapter navigation structure, the navigation structure comprising at least one title and each title comprising at least one chapter; extracting a plurality of keyframes from the video data stream, each keyframe corresponding to one of the titles and chapters and having associated geographic location data; determining a site name for each title and chapter using the geographic location data, wherein the site names for each title have a first level of abstraction, and wherein the site names for each chapter have a second level of abstraction different from the first level of abstraction; and using the site names as a text label for the keyframes corresponding to each title and chapter. - View Dependent Claims (4, 5, 6, 7, 8, 9)
-
-
10. A method for creating a summarization of a video document comprising:
-
organizing video data having associated geographic location data into a tide-and-chapter navigation structure, the navigation structure comprising at least one title and each title comprising at least one chapter; extracting a plurality of keyframes from the video data, each keyframe associated with a corresponding one of the titles and chapters; determining a site name for each of the extracted keyframes using the associated geographic location data, wherein the site names for keyframes associated with the tides have a first level of abstraction, and wherein the site names for keyframes associated with the chapter have a second level of abstraction different from the first level of abstraction; and annotating the extracted keyframes with the corresponding site name. - View Dependent Claims (11, 12, 13, 14, 15)
-
-
16. A video summarization and annotation system comprising:
-
a processor for organizing a video document having geographic location data integrated into the video document into a title-and-chapter navigation structure, the navigation structure comprising at least one tide and each title comprising at least one chapter, extracting a plurality of keyframes from the video document, each keyframe associated with a corresponding one of the titles and chapters, and for mapping a site name for each keyframe using the integrated geographic location data, wherein the site names for keyframes associated with the titles have a first level of abstraction, and wherein the site names for keyframes associated with the chapter have a second level of abstraction different from the first level of abstraction; a data storage system in communication with the processor for storing the extracted keyframes and corresponding site names as a video summary; and a display device for displaying the video summary. - View Dependent Claims (17, 18, 19)
-
-
20. A computer readable medium containing instructions for controlling a computer system to perform a method for annotating a video summarization comprising:
-
organizing video data having associated geographic location data into a title-and-chapter navigation structure, the navigation structure comprising at least one title and each title comprising at least one chapter; extracting a plurality of keyframes from the video data, each keyframe associated with a corresponding one of the titles and chapters; mapping a site name for each of the extracted keyframes using the geographic location data, wherein the site names for keyframes associated with the titles have a first level of abstraction, and wherein the site names for keyframes associated with the chapter have a second level of abstraction different from the first level of abstraction; and
annotating extracted keyframes with the corresponding site name. - View Dependent Claims (21)
-
Specification