Research map / 3D vision / NeRF and radiance fields
Removing Objects From Neural Radiance Fields
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| use confidence-based view selection instead of using all inpainted views (ablation) | NeRF object removal · Synthetic objects | PSNR: 27.568 -> 29.437 | Helped |
| use confidence-based view selection instead of regular 10-frame sampling (ablation) | NeRF object removal · Synthetic objects | PSNR: 27.098 -> 29.437 | Helped |
| use confidence-based view selection instead of regular 50-frame sampling (ablation) | NeRF object removal · Synthetic objects | PSNR: 26.718 -> 29.437 | Helped |
| use confidence-based view selection instead of a single middle frame (ablation) | NeRF object removal · Synthetic objects | PSNR: 26.232 -> 29.437 | Helped |
| use 2D inpainted RGB images with confidence-based selection instead of ignoring masked regions (ablation) | NeRF object removal · Synthetic objects - Masked | PSNR: 26.126 -> 29.437 | Helped |
| add confidence-based selection to training on inpainted RGB and depth (ablation) | NeRF object removal · Synthetic objects - Masked | PSNR: 27.568 -> 29.437 | Helped |
| use predicted depth instead of sensor depth (ablation) | Ours · Synthetic objects - Masked | PSNR: 29.437 -> 26.540 | Hurt |
| use predicted depth instead of sensor depth (ablation) | Ours · Real objects - Masked | PSNR: 25.271 -> 25.010 | Hurt |
| use confidence-based view selection rather than all inpainted views (ablation) | NeRF object removal · Synthetic objects | LPIPS: 0.094 -> 0.078 | Helped |
Closest papers by meaning
- Language-driven Object Fusion into Neural Radiance Fields with Pose-Conditioned Dataset Updates (CVPR 2024 · similarity 0.92)
- InNeRF360: Text-Guided 3D-Consistent Object Inpainting on 360-degree Neural Radiance Fields (CVPR 2024 · similarity 0.91)
- Optimize the Unseen - Fast NeRF Cleanup with Free Space Prior (NeurIPS 2025 · similarity 0.91)
- NeRF On-the-go: Exploiting Uncertainty for Distractor-free NeRFs in the Wild (CVPR 2024 · similarity 0.91)
- Rad-NeRF: Ray-decoupled Training of Neural Radiance Field (NeurIPS 2024 · similarity 0.90)
- InsertNeRF: Instilling Generalizability into NeRF with HyperNet Modules (ICLR 2024 · similarity 0.90)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.