Phase-Functioned Neural Networks for Character Control
Subscribe:  iCal  |  Google Calendar
7, Mar — 12, Jun
London GB   29, May — 1, Jun
Birmingham GB   1, Jun — 4, Jun
Taipei TW   5, Jun — 10, Jun
Los Angeles US   12, Jun — 15, Jun
Latest comments
2 hours ago

Guys! We need "Favorites" tab here on

Great work!

4 hours ago

My motivation wasn't to knock Cem, not as a person nor as a developer. As I said, "this is cool, no doubt about that". I was sharing my personal opinion about the price-point for a material that is so expensive (performance-wise), and pointing out the fact that the same look can be achieved for cheaper (both performance and wallet-wise). I personally find it hard to budget 10s of dollars for a single material, a single effect, etc., but that's me. Other people have money pouring out of their ears and can afford to play like that. The internet is getting less friendly as far as opening dialogues like this. People should be able to have opinions and share them, debate them, without being told to hush up and move along. I hope others buy and use this asset- I'd be curious to see how it stacks up to alternatives out there (again, as I said "I love options"). As far as making my own assets and releasing articles here? It's in the works. And if somebody came along and started a dialogue about issues, opinions they had, or whatever- I would be happy to engage them!

Phase-Functioned Neural Networks for Character Control
4 May, 2017
Daniel Holden, a researcher at Ubisoft Montreal, will be presenting Phase-Functioned Neural Networks for Character Control at this year’s SIGGRAPH. Researcher studies a new kind of neural network called a “Phase-Functioned Neural Network” that can be used to create a character controller suitable for games. This controller would require very little memory, is fast to compute at runtime, and generates high quality motion in many complex situations. 

This paper can really push the boundaries of the industry and change the way characters are controlled. You can watch the video on this idea below:


We present a real-time character control mechanism using a novel neural network architecture called a Phase-Functioned Neural Network. In this network structure, the weights are computed via a cyclic function which uses the phase as an input. Along with the phase, our system takes as input user controls, the previous state of the character, the geometry of the scene, and automatically produces high quality motions that achieve the desired user control. The entire network is trained in an end-to-end fashion on a large dataset composed of locomotion such as walking, running, jumping, and climbing movements fitted into virtual environments. Our system can therefore automatically produce motions where the character adapts to different geometric environments such as walking and running over rough terrain, climbing over large rocks, jumping over obstacles, and crouching under low ceilings. Our network architecture produces higher quality results than time-series autoregressive models such as LSTMs as it deals explicitly with the latent variable of motion relating to the phase. Once trained, our system is also extremely fast and compact, requiring only milliseconds of execution time and a few megabytes of memory, even when trained on gigabytes of motion data. Our work is most appropriate for controlling characters in interactive scenes such as computer games and virtual reality systems.

Read the full paper

Leave a Reply

Be the First to Comment!