Research map / 3D vision / NeRF and radiance fields
Learning Spectral and Polarimetric Clues for One-to-Multimodal Novel View Synthesis
Federico Lincetto, Gianluca Agresti, Mattia Rossi, Piergiorgio Sartor, Pietro Zanuttigh
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 modality-to-luma loss (ablation) | SPoILeR FT · Mono, RGB-only FT | PSNR: 21.04 -> 25.78 | Improved |
| add modality-to-luma loss (ablation) | SPoILeR FT · NIR, RGB-only FT | PSNR: 22.91 -> 26.55 | Improved |
| add modality-to-luma loss (ablation) | SPoILeR FT · MS, RGB-only FT | PSNR: 25.36 -> 25.45 | No clear effect |
| add latent space geometry loss (ablation) | SPoILeR FT · MS, RGB-only FT | PSNR: 23.14 -> 25.45 | Improved |
| add latent space geometry loss (ablation) | SPoILeR FT · Pol, RGB-only FT | PSNR: 23.08 -> 24.25 | Improved |
| add latent space geometry loss (ablation) | SPoILeR FT · RGB, RGB-only FT | PSNR: 29.90 -> 30.07 | No clear effect |
| add inverse function loss (ablation) | SPoILeR FT · NIR, RGB-only FT | PSNR: 25.96 -> 26.55 | Improved |
| add inverse function loss (ablation) | SPoILeR FT · MS, RGB-only FT | PSNR: 25.28 -> 25.45 | No clear effect |
| multi-scene pre-training then RGB-only fine-tuning vs training MMS-FW from scratch | SPoILeR · MMS-DATA RGB | PSNR: 29.53 -> 30.07 | Improved |
| use SPoILeR pre-training instead of MMS-FW with one extra MS/Pol/NIR frame | SPoILeR · MMS-DATA unbalanced modalities, 1 frame | PSNR: gains of ~8, ~7, and ~6 dB | Improved |
| use SPoILeR with 10 extra second-modality frames | SPoILeR · MMS-DATA unbalanced modalities, 10 frames | PSNR: ~2.5 dB on average | Improved |
| render MS directly instead of converting RGB with MST++ then training MMS-FW | SPoILeR · MS rendering | PSNR: 21.99 -> 25.45 | Improved |
Closest papers by meaning
- MultimodalStudio: A Heterogeneous Sensor Dataset and Framework for Neural Rendering across Multiple Imaging Modalities (CVPR 2025 · similarity 0.89)
- NeRF-VPT: Learning Novel View Representations with Neural Radiance Fields via View Prompt Tuning (AAAI 2024 · similarity 0.87)
- SpectralNeRF: Physically Based Spectral Rendering with Neural Radiance Field (AAAI 2024 · similarity 0.86)
- GoLF-NRT: Integrating Global Context and Local Geometry for Few-Shot View Synthesis (CVPR 2025 · similarity 0.86)
- Learning with Unreliability: Fast Few-shot Voxel Radiance Fields with Relative Geometric Consistency (CVPR 2024 · similarity 0.86)
- HyRF: Hybrid Radiance Fields for Memory-efficient and High-quality Novel View Synthesis (NeurIPS 2025 · similarity 0.85)
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.
