Research map / Detection and segmentation / Weakly supervised detection and segmentation
Joint Semantic Mining for Weakly Supervised RGB-D Salient Object Detection
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 spatial semantic modeling (depth-refined pseudo-labels) (ablation) | weakly-supervised saliency backbone · NJUD | F-measure: .627 -> .689 | Helped |
| add textual semantic modeling confidence-weighted label update on top of SSM (ablation) | JSM pipeline · NJUD | F-measure: .689 -> .717 | Helped |
| add textual semantic modeling confidence-weighted label update on top of SSM (ablation) | JSM pipeline · NLPR | MAE: .073 -> .060 | Helped |
| use depth semantics (SSM) instead of self-updating labels (ablation) | pseudo-label update · NJUD | F-measure: .673 -> .689 | Helped |
| use TSM confidence weights instead of equal-weight historical moving average (ablation) | SSM label update · NJUD | F-measure: .698 -> .717 | Helped |
| add CRF to pseudo-labels (ablation) | backbone · NJUD | F-measure: .627 -> .642 | Helped |
| update interval tau=1 instead of 5 (ablation) | JSM · NJUD | F-measure: .717 -> .703 | Hurt |
| update interval tau=10 instead of 5 (ablation) | JSM · NJUD | F-measure: .717 -> .695 | Hurt |
| add SSM in fully supervised setting (ablation) | backbone · NJUD | F-measure: .871 -> .906 | Helped |
| add depth and text semantics vs RGB-based MSW baseline | weakly-supervised SOD · NLPR | MAE: .071 -> .060 | Helped |
Closest papers by meaning
- Promoting Saliency From Depth: Deep Unsupervised RGB-D Saliency Detection (ICLR 2022 · similarity 0.92)
- Railroad Is Not a Train: Saliency As Pseudo-Pixel Supervision for Weakly Supervised Semantic Segmentation (CVPR 2021 · similarity 0.91)
- Weakly Supervised Salient Object Detection Using Image Labels (AAAI 2018 · similarity 0.91)
- Structure-Consistent Weakly Supervised Salient Object Detection with Local Saliency Coherence (AAAI 2021 · similarity 0.90)
- Weakly-Supervised Salient Object Detection Using Point Supervision (AAAI 2022 · similarity 0.90)
- Multi-Source Weak Supervision for Saliency Detection (CVPR 2019 · 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.