Research map / Detection and segmentation / CNN features
CANet: A Context-Aware Network for Shadow Removal
Zipei Chen, Chengjiang Long, Ling Zhang, Chunxia Xiao
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 learned contextual patch matching (CPM) instead of hand-crafted descriptor matching (ablation) | CANet · ISTD whole image | RMSE (LAB): 6.98 -> 6.15 | Improved |
| add learned contextual patch matching (CPM) instead of hand-crafted descriptor matching (ablation) | CANet · SRD whole image | RMSE (LAB): 6.89 -> 5.98 | Improved |
| add CPM instead of MatchNet for patch matching (ablation) | CANet · ISTD whole image | RMSE (LAB): 6.52 -> 6.15 | Improved |
| add CPM instead of MatchNet for patch matching (ablation) | CANet · SRD whole image | RMSE (LAB): 6.35 -> 5.98 | Improved |
| add contextual feature transfer (CFT) (ablation) | CANet · ISTD whole image | RMSE (LAB): 7.54 -> 6.15 | Improved |
| add contextual feature transfer (CFT) (ablation) | CANet · SRD whole image | RMSE (LAB): 6.96 -> 5.98 | Improved |
| add Gaussian sampling to feature transfer instead of direct replacement (ablation) | CANet · ISTD whole image | RMSE (LAB): 6.56 -> 6.15 | Improved |
| add Gaussian sampling to feature transfer instead of direct replacement (ablation) | CANet · SRD whole image | RMSE (LAB): 6.25 -> 5.98 | Improved |
| add context-aware Stage-I with CPM and CFT to one-stage DenseUNet (ablation) | DenseUNet · ISTD whole image | RMSE (LAB): 7.58 -> 6.15 | Improved |
| add context-aware Stage-I with CPM and CFT to one-stage DenseUNet (ablation) | DenseUNet · SRD whole image | RMSE (LAB): 7.28 -> 5.98 | Improved |
| add Gaussian sampling to feature transfer instead of direct replacement (ablation) | CANet · ISTD shadow area | RMSE (LAB): 9.15 -> 8.86 | Improved |
| add CPM instead of MatchNet for patch matching (ablation) | CANet · ISTD shadow area | RMSE (LAB): 9.16 -> 8.86 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- DeshadowNet — DeshadowNet: A Multi-Context Embedding Deep Network for Shadow Removal (CVPR 2017) · better on 2 of 2 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Towards Ghost-Free Shadow Removal via Dual Hierarchical Aggregation Network and Shadow Matting GAN (AAAI 2020 · similarity 0.88)
- DeS3: Adaptive Attention-Driven Self and Soft Shadow Removal Using ViT Similarity (AAAI 2024 · similarity 0.87)
- Recasting Regional Lighting for Shadow Removal (AAAI 2024 · similarity 0.85)
- OmniSR: Shadow Removal Under Direct and Indirect Lighting (AAAI 2025 · similarity 0.85)
- Detail-Preserving Latent Diffusion for Stable Shadow Removal (CVPR 2025 · similarity 0.85)
- Shadow-Enlightened Image Outpainting (CVPR 2024 · 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.
