Research map / Detection and segmentation / CNN features
Distraction-Aware Shadow Detection
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 explicit false-positive and false-negative distraction modeling and distraction-aware loss (ablation) | DSDNet · SBU | BER: 4.42 -> 3.45 | Helped |
| Add explicit false-positive and false-negative distraction modeling and distraction-aware loss (ablation) | DSDNet · UCF | BER: 8.50 -> 7.59 | Helped |
| Add explicit false-positive and false-negative distraction modeling and distraction-aware loss (ablation) | DSDNet · ISTD | BER: 3.41 -> 2.17 | Helped |
| Add the DS module (ablation) | DSDNet · SBU | BER: 3.89 -> 3.45 | Helped |
| Add the DS loss (ablation) | DSDNet · SBU | BER: 3.62 -> 3.45 | Helped |
| Add the FN sub-module (ablation) | DSDNet · SBU | BER: 3.71 -> 3.45 | Helped |
| Add the FP sub-module (ablation) | DSDNet · SBU | BER: 3.68 -> 3.45 | Helped |
| Use distinct FN attention and FP context-convolution strategies rather than simple fusion (ablation) | DSDNet · SBU | BER: 3.79 -> 3.45 | Helped |
| Use attention for FN and context convolution for FP instead of attention for both (ablation) | DSDNet · UCF | BER: 8.41 -> 7.59 | Helped |
| Use attention for FN and context convolution for FP instead of convolution for both (ablation) | DSDNet · ISTD | BER: 2.49 -> 2.17 | Helped |
| Generate distraction supervision from multiple existing models rather than none (ablation) | DSDNet · SBU | BER: 3.84 -> 3.45 | Helped |
| Generate distraction supervision from multiple existing models rather than one model (ablation) | DSDNet · UCF | BER: 8.73 -> 7.59 | Helped |
Closest papers by meaning
- DTTNet: Improving Video Shadow Detection via Dark-Aware Guidance and Tokenized Temporal Modeling (AAAI 2026 · similarity 0.88)
- Triple-Cooperative Video Shadow Detection (CVPR 2021 · similarity 0.87)
- DeS3: Adaptive Attention-Driven Self and Soft Shadow Removal Using ViT Similarity (AAAI 2024 · similarity 0.83)
- MetaShadow: Object-Centered Shadow Detection, Removal, and Synthesis (CVPR 2025 · similarity 0.83)
- When Shadow Removal Meets Intrinsic Image Decomposition: A Joint Learning Framework Using Unpaired Data (AAAI 2025 · similarity 0.81)
- Detail-Preserving Latent Diffusion for Stable Shadow Removal (CVPR 2025 · similarity 0.81)
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.