US 12,277,801 B2
Gesture recognition method, device and computer-readable storage medium
Chenghao Qian, Shenzhen (CN); Miaochen Guo, Shenzhen (CN); Jun Cheng, Shenzhen (CN); and Jianxin Pang, Shenzhen (CN)
Assigned to UBTECH ROBOTICS CORP LTD, Shenzhen (CN)
Filed by UBTECH ROBOTICS CORP LTD, Shenzhen (CN)
Filed on Dec. 31, 2021, as Appl. No. 17/566,734.
Application 17/566,734 is a continuation of application No. PCT/CN2020/135081, filed on Dec. 10, 2020.
Prior Publication US 2022/0189208 A1, Jun. 16, 2022
Int. Cl. G06T 7/73 (2017.01); G06F 3/01 (2006.01); G06T 7/11 (2017.01); G06T 7/55 (2017.01); G06T 11/20 (2006.01); G06V 10/82 (2022.01); G06V 40/20 (2022.01)
CPC G06V 40/20 (2022.01) [G06F 3/017 (2013.01); G06T 11/20 (2013.01); G06V 10/82 (2022.01); G06T 2210/12 (2013.01)] 17 Claims
OG exemplary drawing
 
1. A computer-implemented gesture recognition method, comprising:
acquiring a target image containing a gesture to be recognized;
inputting the target image to a gesture recognition model that comprises a first sub-model, a second sub-model, and a third sub-model that are in a parallel relationship, the first sub-model configured to determine a gesture category and a gesture center point, the second sub-model configured to determine an offset of the gesture center point, and the third sub-model configured to determine a length and a width of a bounding box for the gesture to be recognized, wherein the offset of the gesture center point is caused by the gesture recognition model processing the target image, and wherein the target image is inputted to the first sub-model, the second sub-model and the third sub-model, and the first sub-model, the second sub-model and the third sub-model process the target image in parallel;
acquiring an output result from the gesture recognition model, wherein the output result comprises the gesture category, the gesture center point, and the offset of the gesture center point, and the length and the width of the bounding box; and
determining the gesture category and a position of the bounding box of the gesture to be recognized according to the output result;
wherein the gesture center point and the length and the width of the bounding box are configured for determining a position of a gesture. determining the gesture category and the position of the bounding box of the gesture to be recognized according to the output result comprises:
calculating a target gesture center point according to the gesture center point, and the offset of the gesture center point; and
calculating the position of the bounding box according to the target gesture center point and the length and the width of the bounding box, wherein calculating the position of the bounding box comprises: calculating coordinates of four corners of the bounding box, or calculating coordinates of lower left corner and upper right corner of the bounding box.