Meta & UCL Open-Sources ActionMesh: Fast Video-to-3D Animated Meshes
tom_doerr · x · 2026-08-02
Meta Reality Labs, SpAItial, and University College London (UCL) have open-sourced ActionMesh, a CVPR 2026 project. The model leverages temporal 3D diffusion to generate animated 3D meshes directly from videos.
Focusing on unprecedented quality and speed, ActionMesh can produce 3D mesh assets seamlessly importable into standard 3D software in under a minute. The release also includes actionbench, a new benchmark for evaluating related 3D generation tasks.
More from Multimodal
- Describe your dream world to an AI dragon, which generates the planet for you — repligate · 2026-08-24
- Using kintsugi texture to fix cracks in edited 3D meshes — repligate · 2026-08-24
- Generating Hannibal Character Videos with FL2VA Model — Nimblecloud13 · 2026-08-24
- MiniMax H3 Revives Medieval Short Stories: Complete Workflow Shared — zanatas · 2026-08-24
- NAPE Audio Pretraining Achieves SOTA Without Decoders — kastnerkyle · 2026-08-24
- H3 excels at generating complex space scenes — SIR_NVAX_A_LOT · 2026-08-24