3DTexMOR: 3D Gaussian Multi-Object Removal via Texture-Space Inpainting
We propose 3D Gaussian Multi-Object Removal via Texture-Space Inpainting (3DTexMOR).
ProofPaper ↗
Key points
- 3D object removal aims to remove target objects from reconstructed scenes and complete the geometry and appearance of occluded regions.
- Existing NeRF- and 3DGS-based methods typically inpaint 2D images to guide 3D completion.
- Inconsistent completions across views also introduce conflicting supervision and blurry reconstructions.
- We further introduce geometrically regularized Gaussian completion to constrain the geometry of the completed regions.
Sources (1)
- [1]3DTexMOR: 3D Gaussian Multi-Object Removal via Texture-Space InpaintingarXiv (AI, ML, NLP, CV, robotics, multi-agent) · Oct 8, 04:02 AM
We propose 3D Gaussian Multi-Object Removal via Texture-Space Inpainting (3DTexMOR).
3D object removal aims to remove target objects from reconstructed scenes and complete the geometry and appearance of occluded regions.
Extractive summary: sentences quoted from the sources.