# US 2021/0065384 A1 · Target tracking method, device, system and non-transitory computer readable storage medium

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2021/0065384 A1
- Document ID: US20210065384A1
- Status: Published application
- Kind: A1
- Application number: 16/915,131
- Filed: 2020-06-29
- Published: 2021-03-04
- Earliest claimed: 2019-08-29
- Assignees: Boe Technology Group Co., Ltd.
- Applicants: BOE Technology Group Co., Ltd.
- Inventors: Yuhai Sun; Yu Gu; Jiashuang Zong
- Web page: https://patentgrep.com/patent/US20210065384A1

## CPC

- G06T 7/246: using feature-based methods, e.g. the tracking of corners or segments
- G06F 18/22: Matching criteria, e.g. proximity measures
- G06T 7/248: {involving reference images or patches}
- G06T 7/277: involving stochastic approaches, e.g. using Kalman filters
- G06V 10/757: {Matching configurations of points or features}
- G06V 10/761: {Proximity, similarity or dissimilarity measures}
- G06V 40/103: {Static body considered as a whole, e.g. static pedestrian or occupant recognition}
- G06V 40/23: {Recognition of whole body movements, e.g. for sport training}
- G06T 2207/10016: Video; Image sequence
- G06T 2207/30196: Human being; Person
- G06T 2207/30241: Trajectory
- G06V 10/759: {Region-based matching}
- G06V 2201/07: Target detection
- G06T: IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)
- G06V: IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING

## Abstract

The present disclosure relates to a target tracking method, device, system, and a non-transitory computer-readable storage medium. The method includes: determining an area where a current target is located by performing target detection on a current frame image; extracting a current position information of the area where the current target is located; predicting a position information of an area where each historical target is located at a corresponding moment of the current frame image based on historical position information of an area where each historical target is located in one or more historical frame images; determining a historical target that is the same target as the current target based on a position difference between the current position information and the predicted position information, and tracking the current target.

## Claims

### Claim 1

A target tracking method, comprising:
determining an area where a current target is located, by performing target detection on a current frame image;
extracting a current position information of the area where the current target is located;
predicting a position information of an area where each historical target is located at a moment of the current frame image, based on historical position information of an area where each historical target is located on one or more historical frame images; and
determining a historical target that is the same target as the current target, based on a position difference between the current position information and the predicted position information, and tracking the current target.

### Claim 2 (depends on 1)

The target tracking method according to claim 1, wherein the predicting a position information of an area where each historical target is located at a corresponding moment of the current frame image comprises:
predicting the position information of the area where each historical target is located at the corresponding moment of the current frame image, by a non-linear filtering method.

### Claim 3 (depends on 2)

The target tracking method according to claim 2, wherein the predicting the position information of an area where each historical target is located at the corresponding moment of the current frame image comprises:
predicting the position information of the area where each historical target is located at the corresponding moment of the current frame image, with center point coordinates, aspect ratio and height of the area where each historical target is located on one or more historical frame images, as well as a rate of change in the center point coordinates, a rate of change in the aspect ratio and a rate of change in the height as inputs of the non-linear filtering method.

### Claim 4 (depends on 1)

The target tracking method according to claim 1, further comprising:
extracting a current feature vector of the area where the current target is located on the current frame image;
obtaining a historical feature vector of the area where each historical target is located on one or more historical frame images; and
determining a feature difference between the current target and the each historical target, based on the current feature vector and each historical feature vector;
wherein the determining a historical target that is the same target as the current target comprises:
determining the historical target that is the same target as the current target, based on the feature difference and the position difference.

### Claim 5 (depends on 4)

The target tracking method according to claim 4, wherein the determining a feature difference between the current target and the each historical target comprises:
determining the feature difference, based on a minimum cosine distance between the current feature vector and the each historical feature vector.

### Claim 6 (depends on 1)

The target tracking method according to claim 1, wherein the position difference is determined based on a Mahalanobis distance between the current position information and the predicted position information.

### Claim 7 (depends on 1)

The target tracking method according to claim 1, wherein the tracking the current target comprises:
updating a historical trajectory where the historical target is located, based on the current position information, and forming a tracking trajectory of the current target.

### Claim 8 (depends on 7)

The target tracking method according to claim 7, wherein the forming a tracking trajectory of the current target comprises:
converting coordinates of each trajectory point in the historical trajectory updated in an image coordinate system into coordinates in a world coordinate system, and determining the tracking trajectory.

### Claim 9

A target tracking device, comprising one or more processors configured to:
determine an area where a current target is located, by performing target detection on a current frame image;
extract a current position information of the area where the current target is located;
predict a position information of an area where each historical target is located at a moment of the current frame image, based on historical position information of an area where each historical target is located on one or more historical frame images; and
determine a historical target that is the same target as the current target, based on a position difference between the current position information and the predicted position information, and tracking the current target.

### Claim 10 (depends on 9)

The target tracking device according to claim 9, wherein the predicting a position information of an area where each historical target is located at a corresponding moment of the current frame image comprises:
predicting the position information of the area where each historical target is located at the corresponding moment of the current frame image, by a non-linear filtering method.

### Claim 11 (depends on 10)

The target tracking device according to claim 10, wherein the predicting the position information of an area where each historical target is located at the corresponding moment of the current frame image comprises:
predicting the position information of the area where each historical target is located at the corresponding moment of the current frame image, with center point coordinates, aspect ratio and height of the area where each historical target is located on one or more historical frame images, as well as a rate of change in the center point coordinates, a rate of change in the aspect ratio and a rate of change in the height as inputs of the non-linear filtering method.

### Claim 12 (depends on 9)

The target tracking device according to claim 9, wherein the processor is further configured to:
extract a current feature vector of the area where the current target is located on the current frame image;
obtain a historical feature vector of the area where each historical target is located on one or more historical frame images; and
determine a feature difference between the current target and the each historical target, based on the current feature vector and each historical feature vector;
wherein the determining a historical target that is the same target as the current target comprises:
determine the historical target that is the same target as the current target, based on the feature difference and the position difference.

### Claim 13 (depends on 9)

The target tracking method according to claim 9, wherein the tracking the current target comprises:
updating a historical trajectory where the historical target is located, based on the current position information, and forming a tracking trajectory of the current target.

### Claim 14 (depends on 13)

The target tracking method according to claim 13, wherein the forming a tracking trajectory of the current target comprises:
converting coordinates of each trajectory point in the historical trajectory updated in an image coordinate system into coordinates in a world coordinate system, and determining the tracking trajectory.

### Claim 15 (depends on 1)

A target tracking device, comprising:
a memory; and
a processor coupled to the memory, wherein the processor is configured to perform the target tracking method according to claim 1 based on instructions stored in the memory device.

### Claim 16 (depends on 1)

A non-transitory computer-readable storage medium on which a computer program is stored, which when executed by a processor implements the target tracking method according to claim 1.

### Claim 17 (depends on 9)

A target tracking system, comprising:
the target tracking device according to claim 9.

### Claim 18 (depends on 17)

The target tracking system according to claim 17, further comprising:
one or more image sensors for obtaining a frame image containing a target.

### Claim 19 (depends on 17)

The target tracking system according to claim 17, further comprising:
a display device for displaying the tracking trajectory under a world coordinate system.
