Omni-3DEdit: Generalized Versatile 3D Editing in One-Pass
Paper • 2603.17841 • Published
3D Perception, Reconstruction and Generation
This collection gathers VCLab's research to sense, reconstruct, synthesize, and edit high-fidelity 3D worlds.
Note [CVPR 2026 Highlight] Generalized one-pass 3D editing. | Code: https://github.com/mt-cly/Omni3DEdit
Photo3D: Advancing Photorealistic 3D Generation through Structure-Aligned Detail Enhancement
Paper • 2512.08535 • PublishedNote [CVPR 2026] Photorealistic 3D gen with structure-aligned details. | Code: https://github.com/Liangsanzhu/Photo3D/
One2Scene: Geometric Consistent Explorable 3D Scene Generation from a Single Image
Paper • 2602.19766 • PublishedNote [ICLR 2026] Single-image explorable 3D scene generation. | Code: https://one2scene5406.github.io/
Fast Multi-view Consistent 3D Editing with Video Priors
Paper • 2511.23172 • PublishedNote [AAAI 2026] Multi-view 3D editing with video priors. | Code: https://mt-cly.github.io/ViP3DE/