Research map / Detection and segmentation / CNN features
Pyramid Grafting Network for One-Stage High Resolution 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 |
|---|---|---|---|
| Use the one-stage PGNet instead of DHQSOD-DH | High-resolution SOD · HRSOD-TE | max F-measure: .922 -> .945 | Helped |
| Train PGNet on UHRSD-TR + HRSOD-TR rather than DUTS-TR | PGNet · UHRSD-TE | max F-measure: .931 -> .949 | Helped |
| Train PGNet on UHRSD-TR + HRSOD-TR rather than DUTS-TR + HRSOD-TR | PGNet · DUTS-TE | max F-measure: .917 -> .871 | Hurt |
| Add dual-branch grafting and CMGM to a single-backbone baseline (ablation) | ResNet-18 + Swin PGNet · HRSOD-TE | max F-measure: .878 -> .940 | Helped |
| Add Attention Guided Loss to the grafting model (ablation) | PGNet with CMGM · HRSOD-TE | max F-measure: .940 -> .945 | Helped |
| Use graft pair R5-S2 rather than R5-S4 (ablation) | PGNet · HRSOD-TE | max F-measure: .913 -> .945 | Helped |
| Use graft pair R5-S2 rather than R5-S3 (ablation) | PGNet · HRSOD-TE | max F-measure: .939 -> .945 | Helped |
| Use graft pair R5-S2 rather than R5-S1 (ablation) | PGNet · HRSOD-TE | max F-measure: .937 -> .945 | Helped |
| Use graft pair R5-S2 rather than R5-S4 (ablation) | PGNet · UHRSD-TE | max F-measure: .935 -> .949 | Helped |
| Use graft pair R5-S2 rather than R5-S3 (ablation) | PGNet · UHRSD-TE | max F-measure: .947 -> .949 | Helped |
| Use graft pair R5-S2 rather than R5-S1 (ablation) | PGNet · UHRSD-TE | max F-measure: .947 -> .949 | Helped |
Closest papers by meaning
- Salient Object Detection With Pyramid Attention and Salient Edges (CVPR 2019 · similarity 0.89)
- Cascaded Partial Decoder for Fast and Accurate Salient Object Detection (CVPR 2019 · similarity 0.88)
- Non-Local Deep Features for Salient Object Detection (CVPR 2017 · similarity 0.88)
- Samba: A Unified Mamba-based Framework for General Salient Object Detection (CVPR 2025 · similarity 0.85)
- ESNet: Evolution and Succession Network for High-Resolution Salient Object Detection (ICML 2024 · similarity 0.85)
- Exploring Salient Object Detection with Adder Neural Networks (AAAI 2025 · 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.