US 12,450,822 B2
3D modeling based on neural light field
Zeng Huang, Los Angeles, CA (US); Jian Ren, Marina Del Ray, CA (US); Sergey Tulyakov, Santa Monica, CA (US); Menglei Chai, Los Angeles, CA (US); Kyle Olszewski, Los Angeles, CA (US); and Huan Wang, Somerville, MA (US)
Assigned to Snap Inc., Santa Monica, CA (US)
Filed by Snap Inc., Santa Monica, CA (US)
Filed on Apr. 24, 2024, as Appl. No. 18/644,653.
Application 18/644,653 is a continuation of application No. 17/656,778, filed on Mar. 28, 2022, granted, now 12,002,146.
Prior Publication US 2024/0273809 A1, Aug. 15, 2024
This patent is subject to a terminal disclaimer.
Int. Cl. G06T 15/06 (2011.01); G06T 7/00 (2017.01)
CPC G06T 15/06 (2013.01) [G06T 7/97 (2017.01); G06T 2207/20081 (2013.01); G06T 2207/20084 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A method comprising:
receiving, by one or more processors, a set of images representing a first view of a scene;
applying, by the one or more processors, a machine learning model comprising a neural light field network to the set of images to predict pixel values of a target image, the pixel values of the target image being predicted without processing multiple points along a camera ray directed to a given pixel value; and
generating, by the one or more processors, a model of the scene based on the set of images and the target image.