Making computer animation more agile, acrobatic—and realistic

Animation in film and video games is hard to make realistic: each action typically requires creating a separate controller, while deep reinforcement learning has yet to generate realistic human or animal motion. UC Berkeley computer scientists have now developed an algorithm that uses reinforcement learning to generate realistic simulations that can even recover realistically, after tripping, for example. The same algorithm works for 25 acrobatic and dance tricks, with one month of learning per skill.

Making computer animation more agile, acrobatic—and realistic

It's still easy to tell computer-simulated motions from the real thing - on the big screen or in video games, simulated humans and animals often move clumsily, without the rhythm and fluidity ...

Tue 10 Apr 18 from Phys.org

Virtual robots that teach themselves kung fu could revolutionize video games

Machine learning may make it much easier to build complex virtual characters.

Tue 10 Apr 18 from MIT Technology Review

Deep Mimic - A Virtual Stuntman

Although the title says "A Virtual Stuntman" it is quite difficult to know exactly what the use case for this research is. It clearly has an impact. Take a look at the video and see what you ...

Sat 14 Apr 18 from iProgrammer

AI stuntpeople could lead to more realistic video games

Video game developers often turn to motion capture when they want realistic character animations. Mocap isn't very flexible, though, as it's hard to adapt a canned animation ...

Wed 11 Apr 18 from Engadget

Watch a 'virtual stuntman' break dance and perform martial arts in machine learning breakthrough

University of California researchers developed new simulation technology. The technology will make animated characters' movements more realistic.

Wed 11 Apr 18 from Daily Mail

Making computer animation more agile, acrobatic -- and realistic

Animation in film and video games is hard to make realistic: each action typically requires creating a separate controller, while deep reinforcement learning has yet to generate realistic human ...

Tue 10 Apr 18 from ScienceDaily

Making computer animation more agile, acrobatic -- and realistic, Tue 10 Apr 18 from Eurekalert

  • Pages: 1

Bookmark

Bookmark and Share