US 11,935,254 B2
Systems and methods for predicting depth using style transfer
Rui Guo, San Jose, CA (US); Xuewei Qi, Dublin, CA (US); Kentaro Oguchi, Mountain View, CA (US); and Kareem Metwaly, State College, PA (US)
Assigned to Toyota Motor Engineering & Manufacturing North America, Inc., Plano, TX (US)
Filed by Toyota Motor Engineering & Manufacturing North America, Inc., Plano, TX (US)
Filed on Jun. 9, 2021, as Appl. No. 17/342,853.
Prior Publication US 2022/0398757 A1, Dec. 15, 2022
Int. Cl. G06T 7/50 (2017.01); G06N 3/04 (2023.01); G06N 3/045 (2023.01); G06T 7/70 (2017.01)
CPC G06T 7/50 (2017.01) [G06N 3/045 (2023.01); G06T 7/70 (2017.01); G06T 2207/10028 (2013.01); G06T 2207/20081 (2013.01); G06T 2207/20084 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A prediction system for depth prediction comprising:
a processor;
a memory storing instructions that, when executed by the processor, cause the processor to:
encode, by a style model, an input image to identify content information and identify style information, wherein the content information identifies type features about objects within the input image and the content information is separated from the style information that identifies style features about the objects;
decode, by the style model separate from the style features, the content information into an albedo component and a shading component that generates a lighter representation of the input image;
generate, by the style model, a synthetic image using the albedo component and the shading component; and
provide the synthetic image to a depth model.