Target tracking method and system, readable storage medium, and mobile platform

Patent metadata

Field	Value
Publication number	US-11748968-B2
Application number	US-202117562876-A
Country	US
Kind code	B2
Filing date	Dec 27, 2021
Priority date	Jun 28, 2019
Publication date	Sep 5, 2023
Grant date	Sep 5, 2023

Abstract

The present disclosure provides a target tracking method and system, a readable storage medium, and a mobile platform. The method includes: obtaining a user's trigger operation on an operation button, and generating a trigger instruction based on the trigger operation to generate a candidate target box; displaying, based on the trigger operation, the candidate target box in a current frame of picture displayed on a display interface to correspond to the feature portion of the predetermined target; obtaining, based on the displayed candidate target box, a box selection operation performed by the user on the operation button, and generating a box selection instruction based on the box selection operation to generate a tracking target box, where the box selection instruction is used to determine that the candidate target boxes is a tracking target box; and tracking the target based on the tracking target box.

First claim

What is claimed is:

1. A target tracking method, comprising:
obtaining, by a target tracking system, a trigger operation on an operation button to generate a trigger instruction, wherein the trigger instruction enables the target tracking system to operate in a predetermined box selection mode;
generating, by the target tracking system based on the trigger instruction, at least one candidate target box under the box selection mode;
displaying, by the target tracking system, the at least one candidate target box on a predetermined recognition area of a display interface, wherein the display interface is displaying a target in a current frame of picture, and each of the at least one candidate target box corresponds to a feature portion of the target;
obtaining, based on the at least one candidate target box, a box selection operation on the operation button;
generating, based on the box selection operation, a box selection instruction;
generating, based on the box selection instruction, a tracking target box among the at least one candidate target box; and
tracking the target with the tracking target box.

2. The method according to claim 1 , wherein
the box selection mode includes at least one of a first box selection mode, or a second box selection mode;
in the first box selection mode, the recognition area is the display interface itself;
in the second box selection mode, the recognition area is a partial area of the display interface corresponding to a first default target box; and
the first default target box is by default configured to track a target in the first default target box prior to the generating of the tracking target box.

3. The method according to claim 2 , wherein
the trigger operation includes at least one of a first trigger operation or a second trigger operation;
the first trigger operation renders the target tracking system to generate a first trigger instruction to operate in the first box selection mode; and
the second trigger operation renders the target tracking system to generate a second trigger instruction to operate in the second box selection mode.

4. The method according to claim 1 , wherein
the displaying of the at least one candidate target box includes:
operating the box selection mode based on the trigger instruction; and
generating the at least one candidate target box in the recognition area based on the box selection mode.

5. The method according to claim 1 , wherein
the displaying of the at least one candidate target box includes:
obtaining position information of the at least one candidate target box; and
displaying, based on the position information, the at least one candidate target box in the current frame of picture displayed on the display interface.

6. The method according to claim 2 , further comprising, in the first selection mode:
sorting all of the at least one candidate target box based on central positions of the at least one candidate target box;
obtaining a direction selection operation on the operation button to generate an indicated direction;
switching from a currently selected candidate target box to a directed candidate target box based on the indicated direction; and
selecting the tracking target box based on the directed candidate target box.

7. The method according to claim 2 , further comprising, in the first box selection mode:
setting, in the frame of picture displayed on the display interface, a candidate target box that is closest to a center of the current frame of picture among the at least one candidate target box as a second default target box, wherein
the second default target box by default is configured to track a target in the second default target box by default prior to the generating of the tracking target box.

8. The method according to claim 2 , wherein
the first default target box is located in a center area of a lens of a photographing apparatus that shoots the frame of picture.

9. The method according to claim 2 , further comprising:
operating the second box selection mode;
obtaining a first feature portion of the target in the first default target box in the current frame of picture displayed on the display interface; and
adjusting the first default target box to become one of the at least one candidate target box based on the first feature portion.

10. The method according to claim 2 , further comprising:
operating the second box selection mode; and
maintaining the first default target box upon detecting the portion of the target not presenting in the recognition area.

11. The method according to claim 1 , wherein
the operation button includes at least one of a joystick, or a button.

12. The method according to claim 1 , wherein
the feature portion includes at least one of a portion of the target, or a posture of the target.

13. The method according to claim 12 , wherein, the portion of the target includes at least one of a face, a head, a shoulder, or a torso.

14. The method according to claim 1 , wherein
the photographing apparatus and the operation button are disposed on a mobile platform.

15. The method according to claim 14 , wherein
the mobile platform includes at least one of: a handheld gimbal, an unmanned aerial vehicle, a vehicle, a robot, or a boat.

16. The method according to claim 14 , wherein the tracking of the target with the tracking target box includes:
aligning, based on the tracking target box the mobile platform with the target corresponding to the tracking target box to move synchronously with the target, so as to track the target.

17. The method according to claim 1 , wherein
the target is at least one target, and the at least one candidate target box corresponds to the feature portion of the predetermined target.

18. The method according to claim 1 , wherein
the displaying of the at least one candidate target box includes:
displaying, based on the trigger operation, the at least one candidate target box in the current frame of picture displayed on the display interface, wherein
the candidate target box encloses the feature portion of the target.

19. A target tracking system, comprising:
at least one storage medium storing a set of instructions for target tracking; and
at least one processor in communication with the at least one storage medium, wherein during operation, the at least one processor executes the set of instructions to:
obtain, by a target tracking system, a trigger operation on an operation button to generate a trigger instruction, wherein the trigger instruction enables the target tracking system to operate in a predetermined box selection mode;
generate, by the target tracking system based on the trigger instruction, at least one candidate target box under the box selection mode;
display, by the target tracking system, the at least one candidate target box on a predetermined recognition area of a display interface, wherein the display interface is displaying a target in a current frame of picture, and each of the at least one candidate target box corresponds to a feature portion of the target;
obtain, based on the at least one candidate target box, a box selection operation on the operation button;
generate, based on the box selection operation, a box selection instruction;
generate, based on the box selection instruction, a tracking target box among the at least one candidate target box; and
track the target with the tracking target box.

20. A mobile platform, comprising:
a platform body;
at least one o

Assignees

• Sz Dji Technology Co Ltd

Inventors

• Xiong Feng
• Xue Lijun
• Su Tie

Classifications

• G06V10/25 Primary

  Determination of region of interest [ROI] or a volume of interest [VOI] · CPC title

• G06T7/20

  Analysis of motion (motion estimation for coding, decoding, compressing or decompressing digital video signals H04N19/43, H04N19/51) · CPC title

• G06V10/24

  Aligning, centring, orientation detection or correction of the image · CPC title

• H04N23/61

  based on recognised objects · CPC title

• H04N23/62

  Control of parameters via user interfaces · CPC title