Research map / Detection and segmentation / Weakly supervised detection and segmentation
Multi-Source Uncertainty Mining for Deep Unsupervised Saliency 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 |
|---|---|---|---|
| Mine per-label uncertainty from multiple noisy labels and use it to weight SOD training | SODNet with UMNet · DUTS test | Fmax: 78.42 -> 79.87 | Helped |
| Mine per-label uncertainty from multiple noisy labels and use it to weight SOD training | SODNet with UMNet · DUT-OMRON | Fmax: 77.31 -> 78.71 | Helped |
| Mine per-label uncertainty from multiple noisy labels and use it to weight SOD training | SODNet with UMNet · ECSSD | Fmax: 90.07 -> 90.29 | Helped |
| Mine per-label uncertainty from multiple noisy labels and use it to weight SOD training | SODNet with UMNet · HKU-IS | Fmax: 90.20 -> 90.76 | Helped |
| Use multi-source label feature interaction through Merge-and-Split modules rather than independent branches (ablation) | UMNet · DUTS test and DUT-OMRON | Fmax and S-measure | Helped |
| Use separate label branches with feature interaction rather than one branch over concatenated labels (ablation) | UMNet · DUTS test and DUT-OMRON | Fmax and S-measure | Helped |
| Use learned uncertainty selection rather than treating all refined pseudo labels equally (ablation) | SODNet · DUTS test and DUT-OMRON | Fmax and S-measure | Helped |
| Use learned uncertainty selection rather than majority-voting label selection (ablation) | SODNet · DUTS test and DUT-OMRON | Fmax and S-measure | Helped |
| Use multiple pseudo-label sources rather than only RBD labels (ablation) | UMNet and SODNet · DUTS test and DUT-OMRON | Fmax and S-measure | Helped |
| Use multiple pseudo-label sources rather than only DSR labels (ablation) | UMNet and SODNet · DUTS test and DUT-OMRON | Fmax and S-measure | Helped |
| Use multiple pseudo-label sources rather than only MC labels (ablation) | UMNet and SODNet · DUTS test and DUT-OMRON | Fmax and S-measure | Helped |
| Use multiple pseudo-label sources rather than only HS labels (ablation) | UMNet and SODNet · DUTS test and DUT-OMRON | Fmax and S-measure | Helped |
Closest papers by meaning
- Unsupervised Domain Adaptive Salient Object Detection through Uncertainty-Aware Pseudo-Label Learning (AAAI 2022 · similarity 0.91)
- DeepUSPS: Deep Robust Unsupervised Saliency Prediction via Self-supervision (NeurIPS 2019 · similarity 0.90)
- Deep Unsupervised Saliency Detection: A Multiple Noisy Labeling Perspective (CVPR 2018 · similarity 0.89)
- Texture-Guided Saliency Distilling for Unsupervised Salient Object Detection (CVPR 2023 · similarity 0.87)
- SURE: SUrvey REcipes for building reliable and robust deep networks (CVPR 2024 · similarity 0.84)
- Unsupervised Salient Instance Detection (CVPR 2024 · similarity 0.84)
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.