US 12,288,340 B2
System and method for 3D object perception trained from pure synthetic stereo data
Thomas Kollar, San Jose, CA (US); Kevin Stone, Palo Alto, CA (US); Michael Laskey, Oakland, CA (US); and Mark Edward Tjersland, Mountain View, CA (US)
Assigned to TOYOTA RESEARCH INSTITUTE, INC., Los Altos, CA (US); and TOYOTA JIDOSHA KABUSHIKI KAISHA, Aichi-Ken (JP)
Filed by TOYOTA RESEARCH INSTITUTE, INC., Los Altos, CA (US)
Filed on Jun. 13, 2022, as Appl. No. 17/839,201.
Prior Publication US 2023/0401721 A1, Dec. 14, 2023
Int. Cl. G06T 7/174 (2017.01); G06T 7/00 (2017.01)
CPC G06T 7/174 (2017.01) [G06T 7/97 (2017.01); G06T 2210/12 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A method for 3D object perception, the method comprising:
extracting features from each image of a synthetic stereo pair of images;
generating a low-resolution disparity image based on the features extracted from each image of the synthetic stereo pair of images;
predicting, by a trained neural network, a feature map based only on the low-resolution disparity image and only one of a left image or a right image of the synthetic stereo pair of images; and
generating, by a perception prediction head, a perception prediction of a detected 3D object based on the feature map predicted by the trained neural network.