Research map / 3D vision / Depth and stereo
Shape From Polarization for Complex Scenes in the Wild
Check 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 |
|---|---|---|---|
| Add the proposed polarization representation instead of removing polarization information (ablation) | Ours · SPW | Mean angular error: 27.52 -> 17.86 | Helped |
| Add the proposed polarization representation instead of raw polarization channels (ablation) | Ours · SPW | Mean angular error: 21.77 -> 17.86 | Helped |
| Use the proposed polarization representation instead of Kondo et al.'s representation (ablation) | Ours · SPW | Mean angular error: 18.26 -> 17.86 | Helped |
| Use the proposed polarization representation instead of DeepSfP's representation (ablation) | Ours · SPW | Mean angular error: 18.05 -> 17.86 | Helped |
| Add viewing encoding (ablation) | Ours · SPW | Mean angular error: 22.12 -> 17.86 | Helped |
| Use per-pixel viewing directions instead of NeRF positional encoding (ablation) | Ours · SPW | Mean angular error: 20.31 -> 17.86 | Helped |
| Use per-pixel viewing directions instead of normalized coordinates (ablation) | Ours · SPW | Mean angular error: 18.44 -> 17.86 | Helped |
| Add instance normalization (ablation) | Ours · SPW | Mean angular error: 20.74 -> 17.86 | Helped |
| Add self-attention (ablation) | Ours · SPW | Mean angular error: 21.08 -> 17.86 | Helped |
| Use 8 self-attention blocks rather than 0 (ablation) | Ours · SPW | Mean angular error: 21.08 -> 17.76 | Helped |
| Increase self-attention blocks from 8 to 12 (ablation) | Ours · SPW | Mean angular error: 17.76 -> 17.67 | Helped |
| Train on near-field polarization and evaluate on far-field outdoor scenes | Ours · Far-field outdoor scenes | Qualitative generalization | Helped |
Closest papers by meaning
- SfPUEL: Shape from Polarization under Unknown Environment Light (NeurIPS 2024 · similarity 0.91)
- Spectral and Polarization Vision: Spectro-polarimetric Real-world Dataset (CVPR 2024 · similarity 0.87)
- Polarization Wavefront Lidar: Learning Large Scene Reconstruction from Polarized Wavefronts (CVPR 2024 · similarity 0.86)
- GNeRP: Gaussian-guided Neural Reconstruction of Reflective Objects with Noisy Polarization Priors (ICLR 2024 · similarity 0.86)
- RGB-to-Polarization Estimation: A New Task and Benchmark Study (NeurIPS 2025 · similarity 0.86)
- Normal-NeRF: Ambiguity-Robust Normal Estimation for Highly Reflective Scenes (AAAI 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.