ID-V2V goes open source under Apache 2.0: identity-preserving video restyle with relighting at 720p
minchoi · x · 2026-09-08
Eyeline Labs has open-sourced ID-V2V (SIGGRAPH Asia 2026 paper) under Apache 2.0, with code and model weights public today.
- Generates at 720p and supports longer videos by stitching overlapping clips
- Core capability: identity-preserving video-to-video restyling — same actor, same movement, same scene, new style without losing identity
- Includes relighting: new lighting propagates through the whole video, and it doesn't even require the SAM3 preprocessing step
- Positioned as a "shoot first, restyle later" production workflow
Related event: Netflix open-sources ID-V2V video restyling model(4 posts)→
More from Multimodal
- HeyGen side-by-side: two AI avatars of one person, one still uncanny — HeyGen · 2026-09-09
- A 2M-Param L2 Pix2Pix Model Running Real-Time on a 2012 Ivy Bridge CPU — NoenD_i0 · 2026-09-09
- Weaviate shows how to turn a messy creative archive into semantic search without renaming files — philipvollet · 2026-09-09
- Stanford's BulletTime: decoupling frame index from world time in video generation — GordonWetzstein · 2026-09-09
- MiniMax video model's text rendering is hit or miss — users hunt for a reliable fix — bluetimejt · 2026-09-09
- DIY full audiobook with local TTS: Higgs Audio beats VibeVoice, OmniVoice and Piper — HoodWinked69 · 2026-09-09