Research map / Image generation / Image restoration
Unsupervised Learning for Intrinsic Image Decomposition From a Single Image
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 |
|---|---|---|---|
| use USI³D instead of WH18 | ShapeNet · ShapeNet | MSE Avg.: 0.0273 -> 0.0147 | Helped |
| use USI³D instead of LS18 | MPI-Sintel · MPI-Sintel | MSE Avg.: 0.0233 -> 0.0154 | Helped |
| use USI³D instead of MSCR | MPI-Sintel · MPI-Sintel | DSSIM Avg.: 0.1760 -> 0.1635 | Helped |
| use FY18 instead of USI³D | MPI-Sintel · MPI-Sintel | MSE Avg.: 0.0154 -> 0.0064 | Helped |
| use USI³D instead of MUNIT | MIT intrinsic dataset · MIT intrinsic dataset | MSE Avg.: 0.0184 -> 0.0146 | Helped |
| use FY18 instead of USI³D | MIT intrinsic dataset · MIT intrinsic dataset | MSE Avg.: 0.0146 -> 0.0111 | Helped |
| use USI³D instead of LS18 | IIW · IIW test set | WHDR%(mean): 20.30 -> 18.69 | Helped |
| add prior mapping that predicts reflectance and shading prior codes from the input prior code (ablation) | USI³D on ShapeNet · ShapeNet | MSE Avg.: 0.0171 -> 0.0147 | Helped |
| add prior mapping that predicts reflectance and shading prior codes from the input prior code (ablation) | USI³D on ShapeNet · ShapeNet | LMSE Total: 0.0509 -> 0.0465 | Helped |
| add content consistency loss (ablation) | USI³D on ShapeNet · ShapeNet | MSE Avg.: 0.0203 -> 0.0147 | Helped |
| add content consistency loss (ablation) | USI³D on ShapeNet · ShapeNet | LMSE Total: 0.0529 -> 0.0465 | Helped |
| add physical consistency loss (ablation) | USI³D on ShapeNet · ShapeNet | MSE Avg.: 0.0174 -> 0.0147 | Helped |
Closest papers by meaning
- Self-Supervised Intrinsic Image Decomposition (NeurIPS 2017 · similarity 0.93)
- Revisiting Deep Intrinsic Image Decompositions (CVPR 2018 · similarity 0.92)
- When Shadow Removal Meets Intrinsic Image Decomposition: A Joint Learning Framework Using Unpaired Data (AAAI 2025 · similarity 0.88)
- Intrinsic Image Decomposition for Robust Self-supervised Monocular Depth Estimation on Reflective Surfaces (AAAI 2025 · similarity 0.86)
- SSUN-Net: Spatial-Spectral Prior-Aware Unfolding Network for Pan-Sharpening (AAAI 2025 · similarity 0.86)
- ReasonX: MLLM-Guided Intrinsic Image Decomposition (CVPR 2026 · 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.