How Neural Networks Helped Graphics
summary
The talk explains how neural networks were applied to graphics and visual effects production, using examples like large-scale rendering workflows (e.g., Weta Digital’s cluster runs for Avatar) to argue that such heavy compute pipelines belong within HighLoad concerns. It discusses the compute demands (machine-hours) of rendering effects and how AI techniques can assist or optimize those high-load rendering tasks.
AI-written summary · about
chapters (8)
- 0:00Introduction
- 1:15ML in graphics overview
- 7:24Task differences
- 12:02Rendering clouds problem
- 23:05Training the network
- 25:15Results and speed
- 34:04Main idea and apps
- 44:37Q&A
related talks
40mFueling the AI Revolution with Gaming
41mUnveiling the simple math that powers AI
30mMachine Learning & Neural Networks Demystified
47mThe impact of Generative AI in different industries
20mLightning Talk: Beyond Polygons: The Future of 3D Graphics with Implicit Models
16mUsing AI and Computer Technologies to Enhance IT Infrastructure
34mCreativity and AI: Deep Neural Nets "Going Wild"
18mAI in VFX