Reshoot-Anything: Self-supervised model for in-the-wild video reshooting
rsasaki0109 · x · 2026-08-19
Reshoot-Anything is a self-supervised model that reshoots dynamic monocular videos under novel camera trajectories. It learns implicit 4D spatiotemporal priors by training on pseudo multi-view triplets extracted from monocular footage. During inference, a 4D point-cloud anchor from the source video provides geometric conditioning for the new trajectory. The repo includes code, visualizers, and pipelines.
More from Multimodal
- Luma AI Releases Abra: Diffusion Models Need 10x More Data than LLMs for Scaling — iScienceLuvr · 2026-08-19
- Seeking open-weight models for character reference images like Nano Banana — Odd_Lavishness2236 · 2026-08-19
- MiniMax Design Acts Like a Filmmaking Agent, Breaking Down Scripts — umesh_ai · 2026-08-19
- Music video agent NoSpoon heads to production; dev debates one-shot vs granular control — Kyrannio · 2026-08-19
- Local MiniMax H3 img2video demo turns sketch into photorealistic scene — Important-Radish-722 · 2026-08-19
- Requesting lightweight ComfyUI faceswap workflow — Bossdon01 · 2026-08-19