Research map / Detection and segmentation / Open-set and few-shot detection
A Causal Debiasing Framework for Unsupervised 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 |
|---|---|---|---|
| Apply causal debiasing with DSR pseudo-labels | Strong baseline USOD model · DUTS-TE | Eξ: .843 -> .855 | Helped |
| Apply causal debiasing with DSR pseudo-labels | Strong baseline USOD model · DUTS-TE | Fβ: .747 -> .764 | Helped |
| Apply causal debiasing with DSR pseudo-labels | Strong baseline USOD model · DUTS-TE | M: .070 -> .064 | Helped |
| Add explicit low- and high-level visual contrast fusion (ablation) | Vanilla USOD baseline · DUTS-TE | Eξ: .787 -> .843 | Helped |
| Add causal de-confounding to the contrast fusion block (ablation) | Contrast Fusion Block model · DUTS-TE | Eξ: .843 -> .844 | Helped |
| Add image-level weighting based on spatial distribution bias (ablation) | De-confounded USOD model · DUTS-TE | Eξ: .844 -> .855 | Helped |
| Use res3 rather than res2 as the low-level contrast feature (ablation) | DUTS DCFD with ResNet-50 · DUTS-TE | Eξ: .855 -> .856 | No clear effect |
| Use res5 rather than res4 as the high-level contrast feature (ablation) | DUTS DCFD with ResNet-50 · DUTS-TE | Eξ: .840 -> .855 | Helped |
| Train on DUTS rather than MSRA-B (ablation) | Causal debiasing framework · DUTS-TE | Eξ | Mixed |
| Use DCFD instead of EDNS | Unsupervised salient object detection · DUTS-TE | Eξ: .847 -> .855 | Helped |
Closest papers by meaning
- Size-invariance Matters: Rethinking Metrics and Losses for Imbalanced Multi-object Salient Object Detection (ICML 2024 · similarity 0.88)
- Unbiased Faster R-CNN for Single-source Domain Generalized Object Detection (CVPR 2024 · similarity 0.87)
- DiMSOD: A Diffusion-Based Framework for Multi-Modal Salient Object Detection (AAAI 2025 · similarity 0.86)
- Long-tailed Object Detection Pretraining: Dynamic Rebalancing Contrastive Learning with Dual Reconstruction (NeurIPS 2024 · similarity 0.86)
- UN-DETR: Promoting Objectness Learning via Joint Supervision for Unknown Object Detection (AAAI 2025 · similarity 0.86)
- Unsupervised Salient Instance Detection (CVPR 2024 · similarity 0.86)
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.