Research map / Detection and segmentation / CNN features
Pyramidal Feature Shrinking for 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 |
|---|---|---|---|
| Replace direct feature concatenation with hierarchical adjacent-feature shrinking using PSD (ablation) | PFSNet · DUTS-TE | F-measure: 0.833 -> 0.885 | Helped |
| Add SEM to the direct-concatenation baseline (ablation) | PFSNet · DUTS-TE | F-measure: 0.833 -> 0.880 | Helped |
| Add AFM to PSD (ablation) | PFSNet · DUTS-TE | F-measure: 0.885 -> 0.891 | Helped |
| Add SEM to PSD and AFM (ablation) | PFSNet · DUTS-TE | F-measure: 0.891 -> 0.898 | Helped |
| Set SEM dilated rate D to 2 rather than 1, 3, or 4 (ablation) | PFSNet · ECSSD | F-measure: 0.950 (D=1); 0.952 (D=2); 0.950 (D=3); 0.946 (D=4) | Helped |
| Use adjacent hierarchical feature fusion instead of large-span feature fusion (ablation) | PFSNet · DUTS-TE | F-measure: 0.833 -> 0.885 | Helped |
| Use PFSNet rather than the strongest listed baseline | PFSNet · ECSSD | F-measure: 0.951 -> 0.952 | Helped |
| Use PFSNet rather than the strongest listed baseline | PFSNet · HKU-IS | F-measure: 0.938 -> 0.943 | Helped |
| Use PFSNet rather than the strongest listed baseline | PFSNet · DUTS-TE | F-measure: 0.888 -> 0.898 | Helped |
| Use PFSNet rather than the strongest listed baseline | PFSNet · DUT-OMRON | F-measure: 0.823 -> 0.823 | No clear effect |
| Use PFSNet rather than the strongest listed baseline | PFSNet · PASCAL-S | F-measure: 0.885 -> 0.881 | Hurt |
Closest papers by meaning
- Cascaded Partial Decoder for Fast and Accurate Salient Object Detection (CVPR 2019 · similarity 0.91)
- Structure-Consistent Weakly Supervised Salient Object Detection with Local Saliency Coherence (AAAI 2021 · similarity 0.89)
- Feature Shrinkage Pyramid for Camouflaged Object Detection With Transformers (CVPR 2023 · similarity 0.89)
- Exploring Salient Object Detection with Adder Neural Networks (AAAI 2025 · similarity 0.87)
- DiMSOD: A Diffusion-Based Framework for Multi-Modal Salient Object Detection (AAAI 2025 · similarity 0.86)
- Size-invariance Matters: Rethinking Metrics and Losses for Imbalanced Multi-object Salient Object Detection (ICML 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.