Research map / Detection and segmentation / CNN features
Progressive Feature Polishing Network 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 |
|---|---|---|---|
| Add progressive feature polishing with 2 FPMs (ablation) | PFPN with ResNet-101 · ECSSD | max F: 0.928 -> 0.949 | Helped |
| Add progressive feature polishing with 2 FPMs (ablation) | PFPN with ResNet-101 · DUTS-TE | max F: 0.851 -> 0.888 | Helped |
| Increase the FPM count from 2 to 3 (ablation) | PFPN with ResNet-101 · ECSSD | max F: 0.949 -> 0.950 | Helped |
| Increase the FPM count from 2 to 3 (ablation) | PFPN with ResNet-101 · DUTS-TE | max F: 0.888 -> 0.888 | No clear effect |
| Share weights between the two FPMs instead of using separate weights (ablation) | PFPN with ResNet-101 · ECSSD | max F: 0.949 -> 0.944 | Hurt |
| Share weights between the two FPMs instead of using separate weights (ablation) | PFPN with ResNet-101 · DUTS-TE | max F: 0.888 -> 0.876 | Hurt |
| Add DenseCRF post-processing (ablation) | PFPN · DUTS-TE | max F: 0.888 -> 0.871 | Hurt |
| Use PFPN rather than the strongest reported ResNet baseline | ResNet-based salient object detection · ECSSD | MAE: 0.036 -> 0.033 | Helped |
| Use PFPN rather than the strongest reported ResNet baseline | ResNet-based salient object detection · HKU-IS | max F: 0.928 -> 0.939 | Helped |
| Use PFPN rather than the strongest reported ResNet baseline | ResNet-based salient object detection · DUT-O | max F: 0.805 -> 0.820 | Helped |
| Use PFPN rather than the strongest reported ResNet baseline | ResNet-based salient object detection · DUTS-TE | MAE: 0.043 -> 0.037 | Helped |
| Use PFPN rather than the strongest reported ResNet baseline | ResNet-based salient object detection · PASCAL-S | max F: 0.881 -> 0.892 | Helped |
Closest papers by meaning
- Progressive Attention Guided Recurrent Network for Salient Object Detection (CVPR 2018 · similarity 0.90)
- Recurrently Aggregating Deep Features for Salient Object Detection (AAAI 2018 · similarity 0.89)
- F³Net: Fusion, Feedback and Focus for Salient Object Detection (AAAI 2020 · similarity 0.88)
- Deep Embedding Features for Salient Object Detection (AAAI 2019 · similarity 0.87)
- AugFPN: Improving Multi-Scale Feature Learning for Object Detection (CVPR 2020 · similarity 0.86)
- Pyramidal Feature Shrinking for Salient Object Detection (AAAI 2021 · 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.