Research and Innovate the Future of Interactive Entertainment
summary
EA research describes concrete tools and results: Swish real-time cloth ML replaces hand‑made simulation in shipped games, improving quality; light Stitch 15–20 min captures → ~10 TB per session produces ML-driven photoreal heads from ~100–150 actors; reinforcement‑learning agents train in roughly 24 hours, sometimes discover exploits (“cheat”) and are repurposed for in‑editor playtesting and level difficulty generation; persona mixing wins a Cog award; voice‑to‑face VAE with tongue animation generalizes to unseen speakers and boosts perceived quality; ethical risks include cheating and moderation.
AI-written summary · about
chapters (7)
- 0:00Welcome
- 1:16Machine learning in games
- 10:23Avatar realism with AI
- 12:58Intelligent agents for testing
- 22:40Generative AI for art
- 29:41Academia and industry collaboration
- 30:26Q&A
related talks
01h 01mIntroduction to Practical AI
48mMachine Takeover - Improving Your Game With Machine Learning & AI
42mMaximising Impact with AI: Ethical Strategies for Developing a Top-Rated Game on a Modest Budget
20mStudio Speaks: The Reality of AI Adoption in Games
38mAI in action: How to use practically in iGaming beyond the hype?
29mThe Present & Future of AI and Games
01h 02mart tools: A new era of creativity
24mIEEE CoG 2019 - Industry Day - Christoffer Holmgård (modl.ai)