US 12,112,569 B2
Multi-stage neural network process for keypoint detection in an image
Upal Mahbub, Santee, CA (US); Rakesh Nattoji Rajaram, San Diego, CA (US); and Vasudev Bhaskaran, San Diego, CA (US)
Assigned to QUALCOMM Incorporated, San Diego, CA (US)
Filed by QUALCOMM Incorporated, San Diego, CA (US)
Filed on Nov. 19, 2021, as Appl. No. 17/455,883.
Application 17/455,883 is a continuation in part of application No. 16/700,219, filed on Dec. 2, 2019, granted, now 11,256,956.
Prior Publication US 2022/0076059 A1, Mar. 10, 2022
This patent is subject to a terminal disclaimer.
Int. Cl. G06V 40/16 (2022.01); G06N 3/045 (2023.01); G06T 5/92 (2024.01); G06T 7/11 (2017.01); G06V 40/10 (2022.01)
CPC G06V 40/171 (2022.01) [G06N 3/045 (2023.01); G06T 5/92 (2024.01); G06T 7/11 (2017.01); G06V 40/103 (2022.01)] 30 Claims
OG exemplary drawing
 
1. A method performed by a processor of a computing device for keypoint detection in an image, comprising:
analyzing an image using a first-stage neural network to output a plurality of regions;
analyzing each of the plurality of regions using a respective second-stage neural network to output a plurality of keypoints in each of the plurality of regions; and
analyzing the plurality of keypoints using a third-stage neural network to determine a correction for the plurality of keypoints to provide corrected keypoints.