US 12,462,474 B2
View dependent three-dimensional morphable models
Ke-Li Cheng, San Diego, CA (US); Anupama S, Chennai (IN); Kuang-Man Huang, Zhubei (TW); Chieh-Ming Kuo, Taoyuan (TW); Avani Rao, Bangalore (IN); Chiranjib Choudhuri, Bangalore (IN); Michel Adib Sarkis, San Diego, CA (US); Ajit Deepak Gupte, Bangalore (IN); and Ning Bi, San Diego, CA (US)
Assigned to QUALCOMM Incorporated, San Diego, CA (US)
Filed by QUALCOMM Incorporated, San Diego, CA (US)
Filed on Jun. 21, 2022, as Appl. No. 17/845,884.
Prior Publication US 2023/0410447 A1, Dec. 21, 2023
Int. Cl. G06T 17/00 (2006.01); G06T 7/50 (2017.01); G06T 7/73 (2017.01); G06T 19/20 (2011.01); G06V 40/16 (2022.01)
CPC G06T 17/00 (2013.01) [G06T 7/50 (2017.01); G06T 7/75 (2017.01); G06T 19/20 (2013.01); G06V 40/171 (2022.01); G06T 2200/08 (2013.01); G06T 2207/20081 (2013.01); G06T 2207/20084 (2013.01); G06T 2207/30201 (2013.01); G06T 2219/2021 (2013.01)] 30 Claims
OG exemplary drawing
 
1. A method of generating a three-dimensional (3D) facial model, the method comprising:
obtaining, by an image input of a machine learning model, at least one input image associated with a face;
generating, by the machine learning model based on the at least one input image associated with the face, a face feature vector, wherein the face feature vector encodes one or more features of the face;
obtaining, by a pose input of the machine learning model, a pose for a 3D facial model associated with the face;
generating, by the machine learning model based on the pose, a pose feature vector, wherein the pose feature vector encodes the pose, and wherein the pose is different from a pose of the face associated with the at least one input image;
combining the face feature vector and the pose feature vector to generate a combined feature vector; and
generating, by a neural network layer of the machine learning model based on the combined feature vector, the 3D facial model associated with the face, wherein one or more parameters associated with a shape component of the 3D facial model are conditioned on the pose.