Research map / Detection and segmentation / CNN features
Reverse Attention for Salient Object Detection
Shuhan Chen, Xiuli Tan, Ben Wang, Xuelong Hu
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 reverse attention to side-output residual learning (ablation) | Ours · DUT-OMRON | max F-measure: 0.762 -> 0.786 | Improved |
| add reverse attention to side-output residual learning (ablation) | Ours · DUT-OMRON | MAE: 0.071 -> 0.062 | Improved |
| add reverse attention to side-output residual learning (ablation) | Ours · PASCAL-S | max F-measure: 0.818 -> 0.834 | Improved |
| add reverse attention to side-output residual learning (ablation) | Ours · PASCAL-S | MAE: 0.106 -> 0.104 | Improved |
| add reverse attention to side-output residual learning (ablation) | Ours · MSRA-B | max F-measure: 0.919 -> 0.931 | Improved |
| add reverse attention to side-output residual learning (ablation) | Ours · six saliency benchmarks | max F-measure and MAE: average 1.4% F-measure gain and 0.5% MAE decrease | Improved |
| set residual depth D=2 instead of D=1 (ablation) | Ours · DUT-OMRON | max F-measure: 0.776 -> 0.786 | Improved |
| set residual depth D=2 instead of D=3 (ablation) | Ours · DUT-OMRON | max F-measure: 0.778 -> 0.786 | Improved |
| set residual depth D=2 instead of D=1 (ablation) | Ours · PASCAL-S | MAE: 0.100 -> 0.104 | Hurt |
| combine more side-output residual features (side-output 5 to side-output 1) (ablation) | Ours · PASCAL-S | max F-measure: 0.817 -> 0.834 | Improved |
| combine more side-output residual features (side-output 5 to side-output 1) (ablation) | Ours · DUT-OMRON | max F-measure: 0.755 -> 0.786 | Improved |
| combine more side-output residual features (side-output 5 to side-output 1) (ablation) | Ours · DUT-OMRON | MAE: 0.071 -> 0.062 | Improved |
Closest papers by meaning
- Salient Object Detection With Pyramid Attention and Salient Edges (CVPR 2019 · similarity 0.92)
- Recurrently Aggregating Deep Features for Salient Object Detection (AAAI 2018 · similarity 0.91)
- Deep Embedding Features for Salient Object Detection (AAAI 2019 · similarity 0.91)
- BASNet: Boundary-Aware Salient Object Detection (CVPR 2019 · similarity 0.91)
- Non-Local Deep Features for Salient Object Detection (CVPR 2017 · similarity 0.91)
- Revisiting Salient Object Detection: Simultaneous Detection, Ranking, and Subitizing of Multiple Salient Objects (CVPR 2018 · similarity 0.91)
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.
