NVIDIA unveils HorizonRelight: sliding-window context fixes lighting jumps in long video relighting
NVIDIAAI · x · 2026-09-11
NVIDIA Research and USC introduce HorizonRelight, tackling lighting inconsistency when diffusion-based video models relight long videos.
- Most diffusion video models are trained on short clips, so chunk-based relighting produces visible lighting jumps between segments
- HorizonRelight propagates target-domain context from one sliding window to the next, letting each section pick up where the last left off
- Result: noticeably more consistent lighting across the full video
The thread also highlights an SFU/IP Paris student team that won AI City Challenge Track 5 at ECCV 2026 by adapting Cosmos 3 Nano for generative traffic video forecasting.
Related event: NVIDIA and USC Release HorizonRelight for Consistent Video Relighting(2 posts)→
More from Multimodal
- Curated collection of 153 GPT Astra prompts showcases one-shot 3D and game generation — TheMoonMidas · 2026-09-11
- DeepSeek V4.1 Flash multimodal: 45T image-text tokens and modality-level load balancing — nrehiew_ · 2026-09-11
- Is local AI video upscaling still broken? Hailuo 768p to 2K/4K求助 — Infinite-Emptiness · 2026-09-11
- Astra Can Compose SNES-Style Chiptunes, Vibe Coders Report — AIandDesign · 2026-09-11
- H3 Max video endpoint teaser blurs the line between real and AI footage — noahsolomon · 2026-09-11
- 14-Year Editor Tests GPT-6 Astra: It Watched 84 Clips, Cut, Scored and Subtitled in DaVinci Resolve — TheMoonMidas · 2026-09-11