Results from our new paper on data-driven physics simulation for games. Produces 300 to 4000 times speedup over standard physics simulations by combining Machine Learning, subspace simulation, and lots of training data! https://t.co/1MeYXnLlUp pic.twitter.com/D5DZvmFIvh
— Daniel Holden (@anorangeduck) July 27, 2019
from Twitter https://twitter.com/anorangeduck
July 27, 2019 at 09:53AM
via IFTTT
No comments:
Post a Comment