Research map / 3D vision / NeRF and radiance fields
Multiscale Tensor Decomposition and Rendering Equation Encoding for View Synthesis
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 |
|---|---|---|---|
| replace single-scale tensor decomposition with multiscale tensor decomposition (ablation) | NRFF · NeRF Synthetic (ship) | PSNR: 31.3 dB to 32.3 dB | Helped |
| increase MTD levels from 2 to 16 (ablation) | NRFF · NeRF Synthetic (ship) | PSNR: 32.3 dB to 33.2 dB | Helped |
| replace position encoding with rendering equation encoding in feature space (ablation) | NRFF MTD · NeRF Synthetic | PSNR: 34.58 -> 35.02 | Helped |
| use IDE for view-direction encoding (ablation) | NRFF MTD · NeRF Synthetic | PSNR: 34.58 -> 34.61 | No clear effect |
| encode the rendering equation in color space rather than feature space (ablation) | NRFF MTD · NeRF Synthetic | PSNR: 35.02 -> 33.53 | Hurt |
| increase TensoRF MLP depth to 10 layers (ablation) | TensoRF · NeRF Synthetic | PSNR: 33.43 -> 34.10 | Helped |
| reduce TensoRF learnable features to 13.4M (ablation) | TensoRF · NeRF Synthetic | PSNR: 34.10 -> 33.99 | Hurt |
| use MTD instead of single-scale tensor decomposition (ablation) | NRFF MTD + PE · NeRF Synthetic | PSNR | Helped |
| use multiscale representation (ablation) | NRFF · NeRF Synthetic (ship) | PSNR | Helped |
Closest papers by meaning
- HyRF: Hybrid Radiance Fields for Memory-efficient and High-quality Novel View Synthesis (NeurIPS 2025 · similarity 0.89)
- Learning Robust Generalizable Radiance Field with Visibility and Feature Augmented Point Representation (ICLR 2024 · similarity 0.89)
- GoLF-NRT: Integrating Global Context and Local Geometry for Few-Shot View Synthesis (CVPR 2025 · similarity 0.88)
- NeRFCodec: Neural Feature Compression Meets Neural Radiance Fields for Memory-Efficient Scene Representation (CVPR 2024 · similarity 0.88)
- NeRF-VPT: Learning Novel View Representations with Neural Radiance Fields via View Prompt Tuning (AAAI 2024 · similarity 0.88)
- Neural NeRF Compression (ICML 2024 · similarity 0.87)
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.