US 11,954,791 B2
Fluid simulations using one or more neural networks
Evgenii Tumanov, Moscow (RU); Dmitry Korobchenko, Moscow (RU); and Alexey Solovey, Los Gatos, CA (US)
Assigned to Nvidia Corporation, Santa Clara, CA (US)
Filed by Nvidia Corporation, Santa Clara, CA (US)
Filed on May 23, 2022, as Appl. No. 17/751,360.
Application 17/751,360 is a continuation of application No. 16/766,166, granted, now 11,341,710, previously published as PCT/US2019/062551, filed on Nov. 21, 2019.
Prior Publication US 2022/0358710 A1, Nov. 10, 2022
Int. Cl. G06T 15/00 (2011.01); G06F 30/25 (2020.01); G06N 3/04 (2023.01); G06N 3/08 (2023.01); G06T 13/20 (2011.01); G06T 15/08 (2011.01)
CPC G06T 15/08 (2013.01) [G06F 30/25 (2020.01); G06N 3/04 (2013.01); G06N 3/08 (2013.01); G06T 13/20 (2013.01); G06T 2210/24 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A computer-implemented method, comprising:
determining, for a set of frames, a simulation to be performed on particles located in a domain, the set of frames including at least a first frame and a second frame;
obtaining, for the first frame, motion information of each particle located in the domain, the motion information including a first position and a first velocity;
determining a final position and a final velocity of each particle in the domain at an end of the first frame based at least in part on the motion information and one or more tensors associated with the particles; and
performing, using one or more processors, the simulation for the first frame and the second frame based at least in part on the final position and the final velocity of each particle.