Research map / 3D vision / NeRF and radiance fields
Compressible-composable NeRF via Rank-residual Decomposition
Jiaxiang Tang, Xiaokang Chen, Jingbo Wang, Gang Zeng
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 rank-residual learning before sort-and-truncate compression (CP, most compressed) (ablation) | CCNeRF CP · LEGO compression | PSNR: 24.97 -> 29.53 | Improved |
| add rank-residual learning (HY-S, most compressed 1.5 MB) (ablation) | CCNeRF HY-S · LEGO compression | PSNR: 23.54 -> 29.30 | Improved |
| add rank-residual learning (HY, 36.5 MB) (ablation) | CCNeRF HY · LEGO compression | PSNR: 26.30 -> 31.94 | Improved |
| add rank-residual learning at full rank (no compression) (ablation) | CCNeRF CP/HY-S/HY · LEGO | PSNR: 30.78 -> 30.55 | Hurt |
| discard MLP renderer, use SH for view dependence (vs TensoRF with MLP) | tensor-decomposition NeRF · Synthetic-NeRF | PSNR: 33.14 -> 32.37 | Hurt |
| SH degree 3 instead of 1 (ablation) | CCNeRF · Materials dataset | PSNR: 28.29 -> 28.99 | Improved |
| SH degree 4 instead of 3 (ablation) | CCNeRF · Materials dataset | PSNR: 28.99 -> 28.65 | Hurt |
| parallel training of all stages without detaching vs sequential stage training (ablation) | rank-residual learning · Chair | PSNR: 34.16 -> 34.37 | Improved |
| do not detach earlier-stage outputs in parallel rank-residual training (ablation) | rank-residual learning · Chair | PSNR: 33.69 -> 34.37 | Improved |
Closest papers by meaning
- Neural NeRF Compression (ICML 2024 · similarity 0.92)
- Rad-NeRF: Ray-decoupled Training of Neural Radiance Field (NeurIPS 2024 · similarity 0.90)
- Mask-Based Modeling for Neural Radiance Fields (ICLR 2024 · similarity 0.90)
- Ced-NeRF: A Compact and Efficient Method for Dynamic Neural Radiance Fields (AAAI 2024 · similarity 0.90)
- A Pre-convolved Representation for Plug-and-Play Neural Illumination Fields (AAAI 2024 · similarity 0.89)
- NeRFCodec: Neural Feature Compression Meets Neural Radiance Fields for Memory-Efficient Scene Representation (CVPR 2024 · similarity 0.89)
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.
