WorldSculpt: Compositional Mesh Reconstruction of Cluttered Scenes from Grounded Video
Muyao Niu · hf · 2026-09-07
WorldSculpt adapts a single-object 3D generative prior to multi-view observations, enabling scalable compositional mesh reconstruction of densely cluttered scenes with severe occlusion.
Instead of holistic reconstruction, the method decomposes a scene into individual objects generated independently and composed into a full world, driven by a grounded video. Useful for 3D content creation and simulation scene building.
More from Multimodal
- Motion-Omni generates real-time full-body co-speech motion aligned with spoken dialogue — PKU1898 · 2026-09-07
- AI restoration strips 50 years of degradation from Apollo Eagle moon footage — Glass_Salamander555 · 2026-09-07
- Codex built and runs a local anime pipeline: SD + LoRA → Wan 2.2 on RTX 5060 Ti — Wonderful_Sample6291 · 2026-09-07
- Dev experiments with audio-synced video effects using MiniMax H3 Max in new app Basic Slop — bennash · 2026-09-07
- Simple one-line prompt yields striking cyberpunk flyover in Grok Imagine 1.5 — tetsuoai · 2026-09-07
- DragonballZ AI Animation Test: Character Sheets Drive Seamless Goku Transformation — Ok-Giraffe-8670 · 2026-09-07