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! http://theorangeduck.com/page/subspace-neural-physics-fast-data-driven-interactive-simulation …pic.twitter.com/D5DZvmFIvh
-
-
Yahu sen niye başka şirketin motoruna sallıyorsun, adamlar machine learning ile geliştirme deniyorlar... Ne yapsınlar R&D yapmak yerine Fortnite skini mi satsınlar?
-
This Tweet is unavailable.
New conversation -
-
-
We got that covered in the Just Cause series :)
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
It’s not that physics for CGI are ever really“right” - meaning that they wbe predictive of actual real world phenomena.
-
many recent SIGGRAPH sim papers are comparing with real-world experiments as validation. CGI sims also tend to be far more complex than engineering sims in terms of scale and interacting phenomena
- 1 more reply
New conversation -
-
-
I always thought fluid simulations (splashes when a character walks through water, etc) would be the first use of neural nets for approximate, real time physics effects. Since they are so terrible compared to the other effects in games.
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
as gamers we have all totally smashed our expectations of physics in games.Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
Deep Reinforcement Learning with Michael Bay movies
, training data too explosive
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.