Research map / Detection and segmentation / Weakly supervised detection and segmentation
Weakly-Supervised Salient Object Detection Using Point Supervision
Read the paperCheck 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 edge-detector supervision to the edge-preserving decoder (ablation) | Point-supervised SOD model · PASCAL-S | Fβmax: 0.8629 -> 0.8636 | Helped |
| Add edge-detector supervision to the edge-preserving decoder (ablation) | Point-supervised SOD model · DUT-OMRON | MAE: 0.0769 -> 0.0717 | Helped |
| Refine first-round pseudo-labels with dense CRF (ablation) | Point-supervised SOD model · DUTS-test | Fβmax: 0.8588 -> 0.8606 | Helped |
| Add NSS to edge and CRF pseudo-label refinement (ablation) | Point-supervised SOD model · DUTS-test | MAE: 0.0463 -> 0.0449 | Helped |
| Add NSS to edge and CRF pseudo-label refinement (ablation) | Point-supervised SOD model · ECSSD | Fβmax: 0.935 -> 0.9359 | Helped |
| Use NSS pseudo-labels without uncertain regions (ablation) | Point-supervised SOD model · DUTS-test | Fβmax: 0.8580 -> 0.8586 | Helped |
| Use NSS pseudo-labels without uncertain regions (ablation) | Point-supervised SOD model · DUTS-test | MAE: 0.0449 -> 0.0457 | Hurt |
| Set adaptive-mask hyperparameter γ to 3 instead of 5 (ablation) | Point-supervised SOD model · DUTS-test | Fβmax: 0.8530 -> 0.8467 | Hurt |
| Set adaptive-mask hyperparameter γ to 4 instead of 5 (ablation) | Point-supervised SOD model · DUTS-test | Fβmax: 0.8530 -> 0.8497 | Hurt |
| Set adaptive-mask hyperparameter γ to 6 instead of 5 (ablation) | Point-supervised SOD model · DUTS-test | Fβmax: 0.8530 -> 0.8480 | Hurt |
| Set adaptive-mask hyperparameter γ to 3 instead of 5 (ablation) | Point-supervised SOD model · ECSSD | Fβmax: 0.9326 -> 0.9299 | Hurt |
| Set adaptive-mask hyperparameter γ to 4 instead of 5 (ablation) | Point-supervised SOD model · ECSSD | Fβmax: 0.9326 -> 0.9304 | Hurt |
Closest papers by meaning
- Structure-Consistent Weakly Supervised Salient Object Detection with Local Saliency Coherence (AAAI 2021 · similarity 0.94)
- Weakly Supervised Salient Object Detection Using Image Labels (AAAI 2018 · similarity 0.93)
- Weakly-Supervised Salient Object Detection via Scribble Annotations (CVPR 2020 · similarity 0.93)
- Weakly Supervised Video Salient Object Detection (CVPR 2021 · similarity 0.91)
- Railroad Is Not a Train: Saliency As Pseudo-Pixel Supervision for Weakly Supervised Semantic Segmentation (CVPR 2021 · similarity 0.90)
- WeakPCSOD: Overcoming the Bias of Box Annotations for Weakly Supervised Point Cloud Salient Object Detection (AAAI 2024 · similarity 0.90)
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.