From March 18 to 22, the international video game developers community will meet in San Francisco for the Game Developers Conference (affectionately known as GDC). No less than 28,000 industry professionals will gather together to exchange ideas, chat and network. Of course, among the 750 conferences, panels and round tables, Ubisoft’s studios are well represented with about twenty speakers from our studios from around the world. Our Montreal studio is pleased to have nine employees speak at this prestigious platform!
Experienced representatives to represent our studio in San Francisco
Jennifer Henry, gameplay programmer, will discuss the complex non-authoritative architecture behind the game in her talk: Back to the Future! Working with Deterministic Simulation in ‘For Honor’ . She will deep dive into the mysterious world of deterministic simulation, by presenting the lessons she learned with her ‘For Honor’ gameplay team over the past six years. This solution uses time travel to not hinder fluidity and responsiveness in a game where eight players try to kill each other simultaneously.
Representing the ALICE studio, Stephen Shaw, technical lead – programation, will talk about the latest motion capture innovations using AI in his session on Rethinking Your Performance Capture Pipeline with Machine Learning. The talk will also explore some new technologies and approaches that are improving ALICE’s capture studio, which all contribute to a fully automated motion capture processing solution.
In The Alchemy and Science of Machine Learning for Games, Yves Jacquier, executive director of our studio’s Production Services will talk about AI’s latest application developed by La Forge. He will also discuss ways to incorporate those new paradigms into video games production.
La Forge will also be represented by Daniel Holden, animation programmer, with his talk: A New Era of Performance Capture with Machine Learning. Daniel will dig into the details of several new performance capture technologies developed at La Forge, including a neural network used to automatically clean motion capture data, an end-to-end solution for capture of facial animation from video, and a machine learning solution, which can generate facial animation from raw audio.
Khronos invited Jean-Noé Morissette, technical architect, to share thoughts and perspectives on working with ecosystem tools and the Vulkan API. In addition, he will discuss the challenges of implementing the API within the AAA game pipeline in his talk: Ubisoft’s Experience Developing with Vulkan.
David Lightbown, technical graphic art director, is co-organising the Tools Tutorial Day, with François Nadeau, data scientist who will present 3D Asset Recognition with Deep Learning. In this talk, François will explain how we can integrate 3D game assets from any Ubisoft studios in an internal asset bank automatically through a Deep Neural Network (DNN). The DNN recognizes and classifies the assets, lowering the human cost to organize and standardize the assets across multiple studios and cultures. A real time saver!
Finally, Olivier Delalleau, data scientist, will discuss the ability to train bots to play games during production, which opens up promising opportunities for automated testing and design assistance in Smart Bots for Better Games: Reinforcement Learning in Production.
If you can’t make it to San Francisco, you can see some free content of the GDC event on Twitch. Check it out and support our Montreal colleagues!