Creating for Vision Pro with AI
summary
Demonstrations show AI-generated 360° environments via Skybox (Stable Diffusion), real-time style transfer using ControlNet and Pix2Pix Turbo, and precise per-object masks from SAM/EfficientSam to turn cars, people or walls into pixel art or Van Gogh oil versions. A Blender geometry-nodes pipeline produces real-time 3D assets textured by AI, exported into Vision Pro scenes; demos run live (with some latency) and claim feasibility for on-device or low-latency cloud rendering, though physical-safety and tracking caveats remain.
AI-written summary · about
chapters (9)
- 0:00Introduction and speakers
- 1:03AI and design in Vision Pro
- 4:00Immersive 3D spaces with AI
- 10:42Spatial style transfer demos
- 21:39Real-time AI style transformation
- 28:20AI generation speed and feasibility
- 33:00AI style transfer workflow
- 35:56Future of AI in Vision Pro
- 37:16Q&A and closing
related talks
01h 02mart tools: A new era of creativity
58mStyle Transfer AI - How Neural Networks Generate Art
29mIntegrating Stable Diffusion Generative AI in Artist Workflows
25mDecoding Algorithmic Artistry: Visual Generative AI
40mFrom Concept to Reality: Navigating Real-time Possibilities with Chaos
01h 15mPre-sets and Post-sets: Poetics and Projections
37mMake it Blend: Bringing VFX Techniques into the Browser
25mDesigning at the speed of AI