Automatic extraction of secondary video streams

Patent metadata

Field	Value
Publication number	US-10929680-B2
Application number	US-202016816002-A
Country	US
Kind code	B2
Filing date	Mar 11, 2020
Priority date	Mar 28, 2006
Publication date	Feb 23, 2021
Grant date	Feb 23, 2021

Abstract

A system and method to automatically generate a secondary video stream based on an incoming primary video stream. The method including performing video analytics on the primary video stream to generate one or more analysis results, detecting the first target of interest using the analysis results, automatically extracting a first secondary video stream that captures at least a portion of a first target of interest and has a field of view smaller than that of the primary video stream, tracking the first target of interest, displaying the first secondary video stream, detecting a second target of interest using the analysis results, automatically adapting the first secondary video stream from the primary video stream to capture a portion of the first and second targets of interest, tracking the second target of interest, and displaying the first secondary stream including the portion of the first and second targets of interest.

First claim

We claim:

1. A system to automatically generate a secondary video stream based on a primary video stream, the system comprising:
a computer configured to:
perform video analytics on the primary video stream to generate one or more analysis results, the one or more analysis results including information relating to a first target of interest;
detect the first target of interest using the analysis results;
automatically extract a first secondary video stream from the primary video stream, the first secondary video stream capturing at least a portion of the first target of interest, the first secondary video stream having a field of view smaller than a field of view of the primary video stream;
track the first target of interest;
display the first secondary video stream including the at least the portion of the first target of interest;
detect a second target of interest using the analysis results;
automatically adapt the first secondary video stream from the primary video stream to capture at least a portion of the first and second targets of interest, the first secondary video stream having a field of view smaller than the field of view of the primary video stream;
track the second target of interest; and
display the first secondary video stream including the at least the portion of each of the first and second targets of interest.

2. The system of claim 1 , wherein the primary video stream is a higher resolution primary video stream, and wherein the computer is configured to perform video analytics by down-sampling the higher resolution primary video stream to obtain a lower resolution primary video stream and the computer is configured to perform video analytics on the lower resolution primary video stream.

3. The system of claim 1 , wherein the primary video stream is a higher resolution primary video stream, and wherein the computer is configured to extract the first secondary video stream from the higher resolution primary video stream.

4. The system of claim 1 , wherein the computer is configured to detect foreground objects by determining regions in the primary video stream containing newly-modified objects.

5. The system of claim 1 , wherein the field of view of the first secondary video stream is a portion of the field of view of the primary video stream.

6. The system of claim 1 , further comprising:
a video camera,
wherein the primary video stream is obtained directly from an image sensor of the video camera.

7. The system of claim 1 , wherein the computer is configured to detect the detected first and second targets of interest in the primary video stream and to determine a first direction of motion of the detected first target of interest and a second direction of motion of the detected second target of interest, in the primary video stream.

8. The system of claim 7 , wherein the computer is configured to automatically expand the first secondary video stream in response to the determined first and second directions of motion.

9. The system of claim 1 , wherein the computer is configured to determine a relationship between the detected first and second targets of interest by performing video analytics on the primary video stream, wherein the first secondary video stream is expanded to include the first and second targets of interest in response to the determined relationship between the detected first and second targets of interest.

10. The system of claim 1 , wherein the computer is configured to expand the first secondary video stream to include first and second targets of interest in response to an analysis of movement of the first and second targets of interest in the performing of video analytics on the primary video stream.

11. The system of claim 1 , wherein the computer is configured to automatically extract the first secondary video stream in order to track the detected first and second targets of interest.

12. The system of claim 1 , wherein the computer is configured to automatically extract the first secondary video stream to keep each of the detected first and second targets of interest at a particular position in the first secondary video stream.

13. The system of claim 1 , wherein the computer is configured to maintain a list of one or more currently active targets of interest.

14. The system of claim 1 , wherein the computer is configured to receive configuration information from a user interface to determine how to extract the first secondary video stream.

15. A computer-implemented method of automatically generating a secondary video stream based on primary video stream, comprising:
performing video analytics on the primary video stream to generate one or more analysis results, the one or more analysis results including information relating to a first target of interest;
detecting the first target of interest using the analysis results;
automatically extracting a first secondary video stream from the primary video stream, the first secondary video stream capturing at least a portion of the first target of interest, the first secondary video stream having a field of view smaller than a field of view of the primary video stream;
tracking the first target of interest;
displaying the first secondary stream including the at least the portion of the first target of interest;
detecting a second target of interest using the analysis results;
automatically adapting the first secondary video stream from the primary video stream to capture at least a portion of the first and second targets of interest, the first secondary video stream having a field of view smaller than the field of view of the primary video stream;
tracking the second target of interest; and
displaying the first secondary video stream including the at least the portion of each of the first and second targets of interest.

16. The computer-implemented method of claim 15 , wherein the primary video stream is a higher resolution primary video stream, the method further comprising:
down-sampling the higher resolution primary video stream to obtain a lower resolution primary video stream; and
performing the video analytics on the lower resolution primary video stream.

17. The computer-implemented method of claim 15 , wherein the primary video stream is a higher resolution primary video stream and wherein the automatically extracting the first secondary video stream from the primary video stream comprises:
extracting the first secondary video stream from the higher resolution primary video stream.

18. The computer-implemented method of claim 15 , further comprising:
detecting foreground objects by determining regions in the primary video stream containing newly-modified objects.

19. The computer-implemented method of claim 15 , wherein the field of view of the first secondary video stream is a portion of the field of view of the primary video stream.

20. The computer-implemented method of claim 15 , wherein the primary video stream is obtained directly from an image sensor of a video camera.

21. The computer-implemented method of claim 15 , wherein the first and second targets of interest are determined in the primary video stream, the method further comprising:
determining a first direction of motion of the detected first target of interest and a second direction of motion of the detected second target of interest, in the primary video stream.

22. The computer-implemented method of claim 21 , further comprising:
automatically expanding the first secondary video str

Assignees

• Avigilon Fortress Corp

Inventors

• Scanlon W Andrew
• Chosak Andrew J
• Clark John I W
• Cutting Robert A
• Lipton Alan J
• Myers Gary W

Classifications

• G06F2218/12

  Classification; Matching · CPC title

• G06V20/46 Primary

  Extracting features or characteristics from the video content, e.g. video fingerprints, representative shots or key frames · CPC title

• G06V20/41

  Higher-level, semantic clustering, classification or understanding of video scenes, e.g. detection, labelling or Markovian modelling of sport events or news items (segmenting video sequences G06V20/49) · CPC title

• G06V20/52

  Surveillance or monitoring of activities, e.g. for recognising suspicious objects (recognising microscopic objects G06V20/69) · CPC title

• G06V20/44

  Event detection · CPC title