US 12,347,010 B2
Single image-based real-time body animation
Egor Nemchinov, London (GB); Sergei Gorbatyuk, London (GB); Aleksandr Mashrabov, Los Angeles, CA (US); Egor Spirin, Saint-Petersburg (RU); Iaroslav Sokolov, Saint-Petersburg (RU); Andrei Smirdin, Saint-Petersburg (RU); and Igor Tukh, Saint-Petersburg (RU)
Assigned to Snap Inc., Santa Monica, CA (US)
Filed by Snap Inc., Santa Monica, CA (US)
Filed on Jun. 27, 2023, as Appl. No. 18/214,538.
Application 18/214,538 is a continuation of application No. 17/695,902, filed on Mar. 16, 2022, granted, now 11,727,617.
Application 17/695,902 is a continuation of application No. 17/062,309, filed on Oct. 2, 2020, granted, now 11,308,676, issued on Apr. 19, 2022.
Application 17/062,309 is a continuation of application No. 16/434,185, filed on Jun. 7, 2019, granted, now 10,839,586, issued on Nov. 17, 2020.
Prior Publication US 2023/0343012 A1, Oct. 26, 2023
This patent is subject to a terminal disclaimer.
Int. Cl. G06T 13/40 (2011.01); G06N 3/08 (2023.01); G06T 3/067 (2024.01); G06T 7/194 (2017.01); G06T 15/04 (2011.01); G06T 17/20 (2006.01)
CPC G06T 13/40 (2013.01) [G06N 3/08 (2013.01); G06T 3/067 (2024.01); G06T 7/194 (2017.01); G06T 15/04 (2013.01); G06T 17/20 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A method for single image-based body animation, the method comprising:
receiving, by a computing device, an input image, the input image including an image of a body of a person;
fitting, by the computing device, a model to the image of the body of the person, wherein the model is configured to:
receive a set of pose parameters corresponding to a pose of the body; and
generate, based on the set of pose parameters, an output image, the output image including an image of the body adopting the pose;
receiving, by the computing device, a further set of pose parameters corresponding to a further pose of the body, wherein the further set of pose parameters is generated using a generic model configured to generate, based on the set of pose parameters, a further output image including a further image of the body adopting the pose, wherein the further image of the body omits at least one feature of the body presented in the output image generated by the model;
providing, by the computing device, the further set of pose parameters to the model to generate the output image of the body adopting the further pose; and
generating, by the computing device and based on the output image, a frame of an output video, the frame of the output video including the output image.