Object tracking apparatus, object tracking system, object tracking method, display control device, object detection device, and computer-readable medium

Patent metadata

Field	Value
Publication number	US-12205376-B2
Application number	US-202318237765-A
Country	US
Kind code	B2
Filing date	Aug 24, 2023
Priority date	Sep 26, 2014
Publication date	Jan 21, 2025
Grant date	Jan 21, 2025

Abstract

An object tracking apparatus, method and computer-readable medium for detecting an object from output information of sensors, tracking the object on a basis of a plurality of detection results, generating tracking information of the object represented in a common coordinate system, outputting the tracking information, and detecting the object on a basis of the tracking information.

First claim

The invention claimed is:

1. An object tracking system comprising:
at least one memory storing instructions; and
at least one processor configured to execute the instructions to perform:
generating, via a first detector implemented by the at least one processor, a first detection result of an object from an image, the first detector detecting the object by using a discriminator which has learned a feature of the object, wherein the first detection result is generated by controlling the first detector to extract the feature of the object;
generating, via a second detector implemented by the at least one processor, a second detection result of the object from the image, the second detector detecting the object based on similarity of an appearance of the object detected on a previous frame image, the image being one of a plurality of frame images arranged sequentially in time, wherein the second detection result is generated by controlling the second detector to use the extracted feature of the object in the previous frame image by the first detector; and
integrating the first detection result and the second detection result.

2. The object tracking system according to claim 1 , wherein the extracted feature of the object is at least any one of a color, a pattern, and a shape of the object.

3. The object tracking system according to claim 1 , wherein the at least one processor is further configured to execute the instructions to perform controlling the second detector to store a feature of a region in which the object is detected and detect the object based on the stored feature of the region in the previous frame image.

4. The object tracking system according to claim 3 , wherein the stored feature of the region is at least any one of a color pattern, a color distribution, an edge distribution and a luminance gradient distribution.

5. The object tracking system according to claim 1 , wherein the at least one processor is further configured to execute the instructions to perform integrating the first detection result and the second detection result by associating the object in the first detection result with the object in the second detection result based on a degree of overlap between an object region in the first detection result and an object region in the second detection result.

6. The object tracking system according to claim 1 , wherein the at least one processor is further configured to execute the instructions to perform:
storing the first detection result temporarily in a memory region; and
integrating the first detection result stored in the memory region and the second detection result.

7. An object tracking method performed by at least one processor and comprising:
generating, via a first detector implemented by the at least one processor, a first detection result of an object from an image, the first detector detecting the object by using a discriminator which has learned a feature of the object, wherein the first detection result is generated by controlling the first detector to extract the feature of the object;
generating, via a second detector implemented by the at least one processor, a second detection result of the object from the image, the second detector detecting the object based on similarity of an appearance of the object detected on a previous frame image, the image being one of a plurality of frame images arranged time sequentially in time, wherein the second detection result is generated by controlling the second detector to use the extracted feature of the object in the previous frame image by the first detector; and
integrating the first detection result and the second detection result.

8. The object tracking method according to claim 7 , wherein the extracted feature of the object is at least any one of a color, a pattern, and a shape of the object.

9. The object tracking method according to claim 7 , further comprising:
controlling the second detector to store a feature of a region in which the object is detected and detect the object based on the stored feature of the region in the previous frame image.

10. The object tracking method according to claim 9 , wherein the stored feature of the region is at least any one of a color pattern, a color distribution, an edge distribution and a luminance gradient distribution.

11. The object tracking method according to claim 7 , further comprising:
integrating the first detection result and the second detection result by associating the object in the first detection result with the object in the second detection result based on a degree of overlap between an object region in the first detection result and an object region in the second detection result.

12. The object tracking method according to claim 7 , further comprising
storing the first detection result temporarily in a memory region; and
integrating the first detection result stored in the memory region and the second detection result.

13. A non-transitory computer-readable medium storing instructions executable by at least one processor to perform processing comprising:
generating, via a first detector implemented by the at least one processor, a first detection result of an object from an image, the first detector detecting the object by using a discriminator which has learned a feature of the object, wherein the first detection result is generated by controlling the first detector to extract the feature of the object;
generating, via a second detector implemented by the at least one processor, a second detection result of the object from the image, the second detector detecting the object based on similarity of an appearance of the object detected on a previous frame image, the image being one of a plurality of frame images arranged sequentially in time, wherein the second detection result is generated by controlling the second detector to use the extracted feature of the object in the previous frame image by the first detector; and
integrating the first detection result and the second detection result.

14. The non-transitory computer-readable medium according to claim 13 , wherein the extracted feature of the object is at least any one of a color, a pattern, and a shape of the object.

15. The non-transitory computer-readable medium according to claim 13 , wherein the processing further comprises:
controlling the second detector to store a feature of a region in which the object is detected and detect the object based on the stored feature of the region in the previous frame image.

16. The non-transitory computer-readable medium according to claim 15 , wherein the stored feature of the region is at least any one of a color pattern, a color distribution, an edge distribution and a luminance gradient distribution.

17. The non-transitory computer-readable medium according to claim 13 , wherein the processing further comprises:
integrating the first detection result and the second detection result by associating the object in the first detection result with the object in the second detection result based on a degree of overlap between an object region in the first detection result and an object region in the second detection result.

Assignees

• Nec Corp

Inventors

• Oami Ryoma
• Miyano Hiroyoshi

Classifications

• G06T7/292 Primary

  Multi-camera tracking · CPC title

• G06T2207/30196

  Human being; Person · CPC title

• G06T2207/30241

  Trajectory · CPC title

• G06T2207/30232

  Surveillance · CPC title

• G06T7/246

  using feature-based methods, e.g. the tracking of corners or segments · CPC title