Google DeepMind open-sources GR3EN for controllable video relighting
jon_barron · x · 2026-07-21
Google DeepMind open-sources GR3EN for controllable scene relighting
GR3EN is a generative relighting system for 3D scenes and real-world videos. It lets users click on lights in a video, change their colors, add new light sources, or turn existing ones off, while preserving consistent shadows, reflections, and interreflections.
Key points:
- Works in the browser with zero setup
- Supports real-world videos, including robotics and egocentric footage
- Released with a demo, paper, code, and project page
- Accepted to SIGGRAPH 2026
The work is attributed to Google DeepMind, Google Research, and the University of Amsterdam.
Related event: Google DeepMind Open-Sources GR3EN for Controllable Video Relighting(2 posts)→
More from Multimodal
- Magnific teases October 28 release: 'We're making it move' — cuenca · 2026-10-27
- Midjourney --sref demo: retro Americana Batman art series — gizakdag · 2026-10-27
- AI video is a new medium, not cinema or animation — and it's perfect for depicting dreams — Kyrannio · 2026-10-27
- Dad turns his toddler's 31-year-old quote into a song via SUNO kids' choir — NathanpmYoung · 2026-10-27
- Band's local-AI music video hits 200k views: 900 generations kept 64, 3-second clips beat face drift — Hwang0500 · 2026-10-27
- Adobe Firefly video editor walkthrough: one-sentence Quick Cut turns 32s into 20s — LudovicCreator · 2026-10-26