Research map / 3D vision / NeRF and radiance fields
Affine-Equivariant Kernel Space Encoding for NeRF Editing
Mikołaj Zieliński, Krzysztof Byrski, Tomasz Szczepanik, Dominik Belter, Przemysław Spurek
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (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 |
|---|---|---|---|
| add view-direction restoration after deformation (ablation) | EKS · NeRF-Synthetic edited scenes, Chair | PSNR: 23.80 -> 26.03 | Improved |
| add view-direction restoration after deformation (ablation) | EKS · NeRF-Synthetic edited scenes, Ship | PSNR: 21.72 -> 24.43 | Improved |
| use RT-GPS Mahalanobis neighbour search instead of Euclidean KNN (ablation) | EKS · NeRF-Synthetic edited scenes, Chair | PSNR: 25.58 -> 26.03 | Improved |
| use RT-GPS Mahalanobis neighbour search instead of Euclidean KNN (ablation) | EKS · NeRF-Synthetic static, Ficus | PSNR: 33.21 -> 33.82 | Improved |
| add hash-grid feature distillation (ablation) | EKS · NeRF-Synthetic edited scenes, Chair | PSNR: 25.98 -> 26.03 | No clear effect |
| add hash-grid feature distillation (ablation) | EKS · NeRF-Synthetic edited scenes, qualitative | visual artifacts | Improved |
| add hash-grid feature distillation (ablation) | EKS · NeRF-Synthetic edited scenes, Ship | PSNR: 24.51 -> 24.43 | Mixed |
| EKS vs prior editable methods after edits | EKS · NeRF-Synthetic edited scenes | PSNR: 24.51 -> 26.03 vs GaMeS on Chair | Improved |
| EKS on static scenes vs editable baselines | EKS · NeRF-Synthetic static | PSNR: 36.13 vs 33.82 on Ficus | No clear effect |
Closest papers by meaning
- Feature 3DGS: Supercharging 3D Gaussian Splatting to Enable Distilled Feature Fields (CVPR 2024 · similarity 0.87)
- Neural NeRF Compression (ICML 2024 · similarity 0.87)
- SC-GS: Sparse-Controlled Gaussian Splatting for Editable Dynamic Scenes (CVPR 2024 · similarity 0.87)
- CodecNeRF: Toward Fast Encoding and Decoding, Compact, and High-quality Novel-view Synthesis (AAAI 2025 · similarity 0.87)
- NeRFCodec: Neural Feature Compression Meets Neural Radiance Fields for Memory-Efficient Scene Representation (CVPR 2024 · similarity 0.87)
- LAENeRF: Local Appearance Editing for Neural Radiance Fields (CVPR 2024 · similarity 0.86)
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.
