Methods and systems for object-recognition and link integration in a composite video stream

Patent metadata

Field	Value
Publication number	US-9396397-B2
Application number	US-201414519715-A
Country	US
Kind code	B2
Filing date	Oct 21, 2014
Priority date	Oct 21, 2014
Publication date	Jul 19, 2016
Grant date	Jul 19, 2016

Abstract

Disclosed herein are methods and systems for object recognition and link integration in a composite video stream. One embodiment takes the form of a process that includes detecting an object of interest in a set of video frames. The process also includes tracking the movements of the detected object of interest across a subset of the video frames in the set of video frames. The process further includes generating a composite video stream from the video frames in the subset. The composite video stream shows the tracked movements of the detected object of interest without showing background data from the video frames in the subset. The process also includes outputting the generated composite video stream.

First claim

What is claimed is:

1. A method including:
detecting an object of interest in a set of a plurality of video frames from one or more video sources;
tracking movements of the detected object of interest across a plurality of subset video frames out of the plurality of video frames less than the plurality of video frames;
generating and storing a new composite video stream from the plurality of subset video frames by removing background data from the plurality of subset video frames and adding links in the composite video stream to corresponding subset video frames that link the subset video frame with background data removed in the composite video stream to the respective video frame from the one or more video sources without its background data removed, the generated composite video stream showing the tracked movements of the detected object of interest across the plurality of subset video frames without showing the removed background data and including links to respective video frames from the one or more video sources without its background data removed; and
outputting the generated composite video stream.

2. The method of claim 1 , wherein the object of interest is a person.

3. The method of claim 1 , wherein the object of interest is a face of a person.

4. The method of claim 3 , wherein detecting the object of interest in the set of video frames comprises using at least one of a facial-detection engine and a facial-recognition engine to detect the object of interest in the set of video frames.

5. The method of claim 1 , wherein the object of interest comprises a feature of a person.

6. The method of claim 1 , wherein the object of interest is a set of multiple objects of interest.

7. The method of claim 1 , wherein the set of video frames includes video frames from multiple different video sources, and wherein detecting the object of interest in the set of video frames comprises detecting the object of interest in video frames from the multiple different video sources.

8. The method of claim 7 , wherein the multiple different video sources comprise multiple different video cameras.

9. The method of claim 7 , wherein at least one of the multiple different video sources is a data store containing previously recorded video.

10. The method of claim 1 , wherein outputting the generated composite video stream comprises outputting the generated composite video stream including visible actionable hyperlinks to the respective plurality of video frames for display on at least one user interface.

11. The method of claim 10 , wherein the visible actionable hyperlinks are shifted to track the movements of the detected object of interest in the composite video stream.

12. The method of claim 1 , wherein storing the generated composite video stream comprises storing the generated composite video stream in at least one data store.

13. The method of claim 1 , further comprising:
identifying a set of attributes of the detected object of interest;
generating an identifier from the identified set of attributes; and
storing the generated identifier in association with the generated composite video stream.

14. The method of claim 13 , wherein storing the generated identifier in association with the generated composite video stream comprises storing the composite video stream in a searchable database of such generated composite video streams, the searchable database being indexed by such generated identifiers.

15. The method of claim 14 , wherein the searchable database is searchable using data masks of the identifiers by which the searchable database is indexed.

16. The method of claim 14 , further comprising receiving a query that includes at least one of an object-of-interest identifier and a data mask of an object-of-interest identifier, and responsively returning search results comprising one or more generated composite videos having associated identifiers that match at least one of an identifier from the query and a data mask from the query.

17. The method of claim 1 , wherein generating the composite video stream comprises including searchable metadata in the composite video stream, the searchable metadata comprising at least one of time data and location data.

18. A system comprising:
a communication interface;
a processor; and
data storage containing instructions executable by the processor for causing the system to carry out a set of functions, the set of functions including:
detecting an object of interest in a set of a plurality of video frames from one or more video sources;
tracking movements of the detected object of interest across a plurality of subset video frames out of the plurality of video frames less than the plurality of video frames;
generating and storing a new composite video stream from the plurality of subset video frames by removing background data from the plurality of subset video frames and adding links in the composite video stream to corresponding subset video frames that link the subset video frame with background data removed in the composite video stream to the respective subset video frame from the one or more video sources without the background data removed, the generated composite video stream showing the tracked movements of the detected object of interest across the plurality of subset video frames without showing the removed background data and including links to respective video frames from the one or more video sources without its background data removed; and
outputting the generated composite video stream.

19. The system of claim 18 , wherein outputting the generated composite video stream comprises outputting the generated composite video stream including visible actionable hyperlinks to the respective plurality of video frames for display on at least one user interface.

20. The system of claim 19 , wherein the visible actionable hyperlinks are shifted to track the movements of the detected object of interest in the composite video stream.

Assignees

• Motorola Solutions Inc
• Motorola Solutions Inc

Inventors

• Sabripour Shervin
• Blanco Alejandro G

Classifications

• G06F17/30823

  Physics · mapped topic

• G06K9/00751

  Physics · mapped topic

• G06T2207/30201

  Face · CPC title

• G06F17/30784

  Physics · mapped topic

• G06K9/00288

  Physics · mapped topic