Research map / Label-efficient and robust learning / Semi-supervised learning
A Multi-Task Mean Teacher for Semi-Supervised 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 shadow-edge supervision (ablation) | shadow-region-only MT-CNN · SBU | BER: 5.28 -> 4.07 | Helped |
| add shadow-edge supervision (ablation) | shadow-region-only MT-CNN · UCF | BER: 9.57 -> 8.09 | Helped |
| add shadow-edge supervision (ablation) | shadow-region-only MT-CNN · ISTD | BER: 2.23 -> 1.8 | Helped |
| add shadow-count supervision (ablation) | shadow-region-only MT-CNN · SBU | BER: 5.28 -> 4.72 | Helped |
| add shadow-count supervision (ablation) | shadow-region-only MT-CNN · UCF | BER: 9.57 -> 9.34 | Helped |
| add shadow-count supervision (ablation) | shadow-region-only MT-CNN · ISTD | BER: 2.23 -> 2.04 | Helped |
| add shadow-edge and shadow-count supervision (ablation) | shadow-region-only MT-CNN · SBU | BER: 5.28 -> 3.61 | Helped |
| add mean-teacher consistency on shadow regions (ablation) | shadow-region-only MT-CNN · SBU | BER: 5.28 -> 4.49 | Helped |
| add mean-teacher consistency on shadow regions (ablation) | shadow-region-only MT-CNN · UCF | BER: 9.57 -> 8.29 | Helped |
| add mean-teacher consistency on shadow regions (ablation) | shadow-region-only MT-CNN · ISTD | BER: 2.23 -> 2.12 | Helped |
| add shadow-edge supervision and mean-teacher consistency (ablation) | shadow-region mean-teacher MT-CNN · ISTD | BER: 2.12 -> 1.75 | Helped |
| remove CRF post-processing (ablation) | MTMT-Net · UCF | BER: 7.47 -> 8.06 | Hurt |
Closest papers by meaning
- Single-Stage Instance Shadow Detection With Bidirectional Relation Learning (CVPR 2021 · similarity 0.86)
- Triple-Cooperative Video Shadow Detection (CVPR 2021 · similarity 0.86)
- Video Shadow Detection via Spatio-Temporal Interpolation Consistency Training (CVPR 2022 · similarity 0.86)
- Noisy Label Recovery for Shadow Detection in Unfamiliar Domains (CVPR 2016 · similarity 0.86)
- Humble Teachers Teach Better Students for Semi-Supervised Object Detection (CVPR 2021 · similarity 0.86)
- Detect Any Mirrors: Boosting Learning Reliability on Large-Scale Unlabeled Data with an Iterative Data Engine (CVPR 2025 · similarity 0.83)
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.