itsmetamike · 4D video reshooting
Reshoot 4D
Reshoot 4D is a set of ComfyUI nodes that lets users change the camera path of an existing video. It builds a 4D capture of the scene (approximate real scale, estimated source camera path, the person's motion), lets the user direct a virtual camera, then has LTX 2.3 render every final frame. Version 0.1.0 adds per-key camera settings (aperture, focus, vignette, grain), import and export of .fbx, .abc and .usd cameras, zoom keys for dolly zooms, and per-shot looks, with finishing at 24 to 60 fps with motion blur, up to 4K. The page lists limitations: no fixed 3D world, so unseen areas are regenerated and can differ between angles, and fast motion can smear faces. It runs locally on a CUDA machine, uses open weights, and the code is AGPL-3.0, with models keeping their own licences.
Open the source ↗Similar Video & Animation releases
- PixVid: Back to Pixels, Lifting Latent Video Diffusion into Pixel Space
Yanke Song, Shuo Chen, Chris Rockwell, Ming-Hsuan Yang, Xun Huang et al. · Pixel-space video diffusion · Week 47, 2026
- Open-H3-Regenerate-2K
NVlabs (SANA / SoL-Engine team) · Open video refiner · Week 47, 2026
- Seedance 2.5 Camera Control (pill pipeline)
Sirio Berati · Code-driven AI camera control · Week 47, 2026
- Gemini Omni 1.2 Flash
Google DeepMind · Video generation model · Week 46, 2026
- Gemini Omni 1.2 Flash
Google DeepMind · Video Generation Model · Week 46, 2026
- SenseNova 6.8 Code2Video
OpenSenseNova (SenseNova) · Code-driven video model · Week 45, 2026
Published by Quanta AI Labs, an AI agency in India. Browse every week in AI News Weekly.