Загрузка видео...

Не удалось загрузить видео

На главную

Whatever happened to training physics based character controllers using RL? This was all the rage a few years ago. Every clip here looks cool af.

29,075 просмотров • 3 месяцев назад •via X (Twitter)

Комментарии: 11

Фото профиля Chen Tessler @ CoRL
Chen Tessler @ CoRL3 месяцев назад

still going strong. Check out our latest work GPC:

Фото профиля AVB
AVB3 месяцев назад

This is really cool 😍

Фото профиля AVB
AVB3 месяцев назад

Papers in this video:

Фото профиля Carlos DP ✈️ CoRL 2026
Carlos DP ✈️ CoRL 20263 месяцев назад

it's still all the rage this year! Check out GPC, SMP, HIL, etc. These are experiments with GPC:

Фото профиля Chandra Shekhar 📟
Chandra Shekhar 📟3 месяцев назад

peopel who were doing this are now busy minting synthetic data en masse

Фото профиля CIX 🦾
CIX 🦾3 месяцев назад

We do this with robots with the @rek sim

Фото профиля pgBouncer
pgBouncer3 месяцев назад

If a virtual object knows how to behave in accurately modeled virtual space, it should know how to in reality Then apply a multiverse simulation to train it on all possibly outcomes Insane potential. Something im very interested in

Фото профиля DK
DK3 месяцев назад

RL slow, RL expensive, RL fails to generalize to edge cases etc

Фото профиля Fajar M Reza
Fajar M Reza3 месяцев назад

Character-control progress may depend more on real-world physics than prettier demonstrations.

Фото профиля İsmail Keskin
İsmail Keskin3 месяцев назад

Probably still eating too much resources. The people doing this research were mostly from AAA studios where resources are already tight. Also unlike traditional animation when this fails it fails in miserable ways. Indies could use this though.

Фото профиля Alec
Alec3 месяцев назад

Been a bit busy, but wanted to jump back here and mention I'm working on exactly this. There's no reason this shouldn't be the future of characters in games.

Похожие видео