quan-meng/seen2scene
Completes realistic 3D scenes from incomplete real-world scans
View on index · View in 3D Map
Related repos
- Compositional scene reconstruction from real videos
- Real-time expressive 3D human mesh recovery
- Single-image 3D scene generation in one pass
- 3D reconstruction from hand-object videos
- Reconstructs 3D humans from photo collections
- Reconstructs 3D scenes from a photograph
- Self-supervised 3D reconstruction from images without labels
- Perceives 3D scenes from videos and photos
- Amodal 3D reconstruction from unposed images
- Builds 3D scene graphs in real-time
