SYSTEMS AND METHODS FOR OBJECT TRACKING

• US 20170053167A1
• Filed: 08/18/2015
• Published: 02/23/2017
• Est. Priority Date: 08/18/2015
• Status: Active Grant

• Current Assignee: Qualcomm, Inc.
• Original Assignee: Qualcomm, Inc.
• Inventors: Ren, Jianfeng, Guo, Feng, Yang, Ruiduo
• Granted Patent: US 10,586,102 B2
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G06T 2207/30241 Trajectory
• CPC Class Codes: G06T 3/40 Scaling of whole images or parts thereof, e.g. expanding or contracting
• CPC Class Codes: G06T 7/246 using feature-based methods, e.g. the tracking of corners or segments

Abstract

A method performed by an electronic device is described. The method includes obtaining a first frame of a scene. The method also includes performing object recognition of at least one object within a first bounding region of the first frame. The method further includes performing object tracking of the at least one object within the first bounding region of the first frame. The method additionally includes determining a second bounding region of a second frame based on the object tracking. The second frame is subsequent to the first frame. The method also includes determining whether the second bounding region is valid based on a predetermined object model.

Claims

1. A method performed by an electronic device, comprising:

obtaining a first frame of a scene;

performing object recognition of at least one object within a first bounding region of the first frame;

performing object tracking of the at least one object within the first bounding region of the first frame;

determining a second bounding region of a second frame based on the object tracking, wherein the second frame is subsequent to the first frame; and

determining whether the second bounding region is valid based on a predetermined object model.

2. The method of claim 1, further comprising performing object detection of the at least one object.

3. The method of claim 2, further comprising obtaining the predetermined object model corresponding to the object.

4. The method of claim 1, further comprising:

extracting one or more features based on the first bounding region; and

updating object tracking based on the one or more features.

5. The method of claim 1, further comprising:

determining whether the object is within the second bounding region if the second bounding region is invalid; and

refining the second bounding region if the object is within the second bounding region.

6. The method of claim 5, wherein refining the second bounding region comprises resizing the second bounding region to exclude one or more pixels that do not correspond to the object.

7. The method of claim 1, further comprising detecting the object in a third frame after the second frame if the object is not within the second bounding region.

8. The method of claim 1, further comprising performing assisted driving based on the object.

9. The method of claim 1, further comprising obtaining the predetermined object model corresponding to the object, wherein obtaining the predetermined object model comprises selecting the predetermined object model from a plurality of predetermined object models.

10. The method of claim 9, wherein at least one of the plurality of predetermined object models comprises a predetermined object model of a traffic sign.

11. An electronic device, comprising:

a processor configured to;

obtain a first frame of a scene;

perform object recognition of at least one object within a first bounding region of the first frame;

perform object tracking of the at least one object within the first bounding region of the first frame;

determine a second bounding region of a second frame based on the object tracking, wherein the second frame is subsequent to the first frame; and

determine whether the second bounding region is valid based on a predetermined object model.

12. The electronic device of claim 11, wherein the processor is configured to perform object detection of the at least one object.

13. The electronic device of claim 12, wherein the processor is configured to obtain the predetermined object model corresponding to the object.

14. The electronic device of claim 11, wherein the processor is configured to:

extract one or more features based on the first bounding region; and

update object tracking based on the one or more features.

15. The electronic device of claim 11, wherein the processor is configured to:

determine whether the object is within the second bounding region if the second bounding region is invalid; and

refine the second bounding region if the object is within the second bounding region.

16. The electronic device of claim 15, wherein the processor is configured to resize the second bounding region to exclude one or more pixels that do not correspond to the object.

17. The electronic device of claim 11, wherein the processor is configured to detect the object in a third frame after the second frame if the object is not within the second bounding region.

18. The electronic device of claim 11, wherein the processor is configured to perform assisted driving based on the object.

19. The electronic device of claim 11, wherein the processor is configured to obtain the predetermined object model corresponding to the object, wherein the processor is configured to obtain select the predetermined object model from a plurality of predetermined object models.

20. The electronic device of claim 19, wherein at least one of the plurality of predetermined object models comprises a predetermined object model of a traffic sign.

21. A computer-program product, comprising a non-transitory tangible computer-readable medium having instructions thereon, the instructions comprising:

code for causing an electronic device to obtain a first frame of a scene;

code for causing the electronic device to perform object recognition of at least one object within a first bounding region of the first frame;

code for causing the electronic device to perform object tracking of the at least one object within the first bounding region of the first frame;

code for causing the electronic device to determine a second bounding region of a second frame based on the object tracking, wherein the second frame is subsequent to the first frame; and

code for causing the electronic device to determine whether the second bounding region is valid based on a predetermined object model.

22. The computer-program product of claim 21, the instructions further comprising code for causing the electronic device to perform object detection of the at least one object.

23. The computer-program product of claim 21, the instructions further comprising:

code for causing the electronic device to determine whether the object is within the second bounding region if the second bounding region is invalid; and

code for causing the electronic device to refine the second bounding region if the object is within the second bounding region.

24. The computer-program product of claim 21, the instructions further comprising code for causing the electronic device to select the predetermined object model from a plurality of predetermined object models.

25. The computer-program product of claim 24, wherein at least one of the plurality of predetermined object models comprises a predetermined object model of a traffic sign.

26. An apparatus, comprising:

means for obtaining a first frame of a scene;

means for performing object recognition of at least one object within a first bounding region of the first frame;

means for performing object tracking of the at least one object within the first bounding region of the first frame;

means for determining a second bounding region of a second frame based on the object tracking, wherein the second frame is subsequent to the first frame; and

means for determining whether the second bounding region is valid based on a predetermined object model.

27. The apparatus of claim 26, further comprising means for performing object detection of the at least one object.

28. The apparatus of claim 26, further comprising:

means for determining whether the object is within the second bounding region if the second bounding region is invalid; and

means for refining the second bounding region if the object is within the second bounding region.

29. The apparatus of claim 26, further comprising means for selecting the predetermined object model from a plurality of predetermined object models.

30. The apparatus of claim 29, wherein at least one of the plurality of predetermined object models comprises a predetermined object model of a traffic sign.