Research map / Detection and segmentation / CNN features
Bidirectional Feature Pyramid Network with Recurrent Attention Residual Modules for Shadow Detection
Lei Zhu, Zijun Deng, Xiaowei Hu, Chi-Wing Fu, Xuemiao Xu, Jing Qin, Pheng-Ann Heng
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 bidirectional pathway to FPN (ablation) | FPN · SBU test | BER: 5.78 -> 5.37 | Improved |
| add bidirectional pathway to FPN (ablation) | FPN · UCF test | BER: 7.70 -> 7.11 | Improved |
| add RAR modules to FPN (ablation) | FPN · SBU test | BER: 5.78 -> 4.33 | Improved |
| add RAR modules to FPN (ablation) | FPN · UCF test | BER: 7.70 -> 6.09 | Improved |
| use RAR modules instead of plain residual modules in the bidirectional network (ablation) | BDRAR · SBU test | BER: 4.23 -> 3.64 | Improved |
| use RAR modules instead of plain residual modules in the bidirectional network (ablation) | BDRAR · UCF test | BER: 6.71 -> 5.30 | Improved |
| add recurrence to attention residual module (2 steps vs 1) (ablation) | BDAR · SBU test | BER: 3.74 -> 3.64 | Improved |
| add recurrence to attention residual module (2 steps vs 1) (ablation) | BDAR · UCF test | BER: 6.19 -> 5.30 | Improved |
| increase recurrent steps from 2 to 3 (ablation) | BDRAR · SBU test | BER: 3.64 -> 3.89 | Hurt |
| share weights across recurrent steps in RAR (ablation) | BDRAR · SBU test | BER: 3.89 -> 3.64 | Improved |
| share weights across recurrent steps in RAR (ablation) | BDRAR · UCF test | BER: 5.66 -> 5.30 | Improved |
| use BDRAR instead of DSC | DSC · SBU test | BER: 5.59 -> 3.64 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.
- DeshadowNet — DeshadowNet: A Multi-Context Embedding Deep Network for Shadow Removal (CVPR 2017) · better on 2 of 2 shared results
- Amulet — Amulet: Aggregating Multi-Level Convolutional Features for Salient Object Detection (ICCV 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
- Direction-Aware Spatial Context Features for Shadow Detection (CVPR 2018 · similarity 0.91)
- Direction-Aware Spatial Context Features for Shadow Detection and Removal (TPAMI 2019 · similarity 0.91)
- Single-Stage Instance Shadow Detection With Bidirectional Relation Learning (CVPR 2021 · similarity 0.88)
- Residual Attention Network for Image Classification (CVPR 2017 · similarity 0.87)
- DeshadowNet: A Multi-Context Embedding Deep Network for Shadow Removal (CVPR 2017 · similarity 0.87)
- Feature Pyramid Networks for Object Detection (CVPR 2017 · similarity 0.87)
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.
