SYSTEM AND METHOD TO IMPROVE OBJECT TRACKING USING TRACKING FINGERPRINTS

• US 20160180546A1
• Filed: 12/19/2014
• Published: 06/23/2016
• Est. Priority Date: 12/19/2014
• Status: Active Grant

• Current Assignee: The Boeing Co.
• Original Assignee: The Boeing Co.
• Inventors: Mundhenk, Terrell N., Kim, Kyungnam, Jeong, Changsoo, Owechko, Yuri
• Granted Patent: US 9,940,726 B2
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G06T 2207/10016 Video; Image sequence
• CPC Class Codes: G06T 7/248 involving reference images or patches
• CPC Class Codes: G06T 7/251 involving models
• CPC Class Codes: G06T 7/292 Multi-camera tracking
• CPC Class Codes: G06T 7/74 involving reference images or patches

Abstract

A method includes receiving image data at a first tracking system. The image data may represent a region in an image of a sequence of images. The method includes generating a first tracking fingerprint based on the image data. The method includes comparing the first tracking fingerprint and a second tracking fingerprint. The method further includes providing an output from the first tracking system to a second tracking system based on a result of the comparison of the first tracking fingerprint and the second tracking fingerprint.

Claims

1. A method comprising:

receiving image data at a first tracking system, wherein the image data represents a region in an image of a sequence of images;

generating a first tracking fingerprint based on the image data;

comparing the first tracking fingerprint and a second tracking fingerprint; and

providing an output from the first tracking system to a second tracking system based on a result of the comparison of the first tracking fingerprint and the second tracking fingerprint.

2. The method of claim 1, wherein the second tracking system is configured to:

track an object in the sequence of images based on an object model, wherein the object model is generated by the second tracking system based on data representing the object in a first image of the sequence of images; and

modify the object model or avoid a modification to the object model based on the output from the first tracking system.

3. The method of claim 1, wherein the first tracking fingerprint is generated by applying a mathematical model to feature data extracted from the image data.

4. The method of claim 1, wherein comparing the first tracking fingerprint and the second tracking fingerprint comprises determining a difference between the first tracking fingerprint and the second tracking fingerprint based on a distance measurement between the first tracking fingerprint and the second tracking fingerprint.

5. The method of claim 1, further comprising:

receiving the image data from a motion detector, wherein the image data includes motion-triggered image data;

determining whether a difference between the first tracking fingerprint and the second tracking fingerprint is less than a first threshold value; and

receiving a confidence value from the second tracking system prior to providing the output, wherein the confidence value indicates a likelihood that the second tracking system is correctly tracking an object.

6. The method of claim 5, further comprising determining whether the confidence value is less than a confidence threshold when the difference between the first tracking fingerprint and the second tracking fingerprint is less than the first threshold value.

7. The method of claim 6, further comprising:

determining whether the second tracking system has lost track of the object when the confidence value is less than the confidence threshold; and

providing a location update instruction and the motion-triggered image data as the output to the second tracking system when the second tracking system has not lost track of the object.

8. The method of claim 7, wherein the second tracking system is configured to, in response to receiving the location update instruction, output location data determined based on the motion-triggered image data.

9. The method of claim 7, further comprising:

determining whether the difference is less than a second threshold value when the second tracking system has lost track of the object; and

providing an object model re-initialization instruction and the motion-triggered image data as the output to the second tracking system when the difference is less than the second threshold value.

10. The method of claim 9, wherein the second tracking system is configured to, in response to receiving the object model re-initialization instruction, modify an object model used by the second tracking system based on the motion-triggered image data.

11. The method of claim 6, further comprising:

determining whether the difference is less than a third threshold value; and

storing the first tracking fingerprint in a memory when the difference is less than the third threshold value, wherein the second tracking fingerprint is stored in the memory.

12. The method of claim 1, further comprising:

receiving the image data from the second tracking system prior to the second tracking system modifying an object model based on the image data; and

wherein comparing the first tracking fingerprint and the second tracking fingerprint comprises determining whether a difference between the first tracking fingerprint and the second tracking fingerprint is less than a fourth threshold value.

13. The method of claim 12, further comprising providing a model update instruction as the output to the second tracking system when the difference is less than the fourth threshold value, wherein the second tracking system is configured to modify the object model based on receipt of the model update instruction.

14. The method of claim 12, further comprising providing a prevent update instruction as the output to the second tracking system when the difference is greater than or equal to the fourth threshold value, wherein the second tracking system is configured to avoid a modification of the object model based on receipt of the prevent update instruction.

15. A tracking system comprising:

one or more processors; and

a memory coupled to the one or more processors, the memory storing instructions that, when executed by the one or more processors, cause the one or more processors to perform operations comprising;

receiving image data representing a region in an image of a sequence of images;

generating a first tracking fingerprint based on the image data;

comparing the first tracking fingerprint and a second tracking fingerprint; and

providing an output to a second tracking system based on a result of the comparison of the first tracking fingerprint and the second tracking fingerprint.

16. The tracking system of claim 15, wherein the sequence of images is received from an image capture device, and wherein the processor, the memory, and the image capture device are integrated in an autonomous vehicle.

17. The tracking system of claim 16, wherein the second tracking system is configured to provide location data to a server, to another system of the autonomous vehicle, or a combination thereof.

18. The tracking system of claim 15, wherein the operations further comprise storing the first tracking fingerprint in the memory based on the result of the comparison, wherein the second tracking fingerprint is stored in the memory.

19. A non-transitory computer readable medium storing instructions that, when executed by a processor, cause the processor to:

receive image data at a first tracking system, wherein the image data represents a region in an image of a sequence of images;

generate a first tracking fingerprint based on the image data;

compare the first tracking fingerprint and a second tracking fingerprint; and

provide an output from the first tracking system to a second tracking system based on a result of the comparison of the first tracking fingerprint and the second tracking fingerprint.

20. The non-transitory computer readable medium of claim 19, wherein the second tracking fingerprint is stored in a memory by the first tracking system, and wherein the output enables the second tracking system to re-acquire an object in a subsequent image of the sequence of images after the object is at least partially occluded in one or more images of the sequence of images.