Research map / Detection and segmentation / CNN features
Where to Focus: Investigating Hierarchical Attention Relationship for Fine-Grained Visual Classification
Yang Liu, Lei Zhou, Pengcheng Zhang, Xiao Bai, Lin Gu, Xiaohan Yu, Jun Zhou, Edwin R. Hancock
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (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 region feature mining (RFM) module to the baseline (ablation) | Baseline · CUB (4 levels) | wAP: 88.9 -> 90.3 | Improved |
| add cross-hierarchical orthogonal fusion (COF) on top of RFM (ablation) | Ours-RF · CUB (4 levels) | wAP: 90.3 -> 91.8 | Improved |
| add region feature mining (RFM) module to the baseline (ablation) | Baseline · Butterfly-200 | wAP: 88.8 -> 90.4 | Improved |
| add cross-hierarchical orthogonal fusion (COF) on top of RFM (ablation) | Ours-RF · Butterfly-200 | wAP: 90.4 -> 91.2 | Improved |
| add region feature mining (RFM) module to the baseline (ablation) | Baseline · VegFru | wAP: 88.7 -> 90.7 | Improved |
| add cross-hierarchical orthogonal fusion (COF) on top of RFM (ablation) | Ours-RF · VegFru | wAP: 90.7 -> 91.4 | Improved |
| add RFM module to the baseline (ablation) | Baseline · FGVC-Aircraft | wAP: 93.0 -> 94.0 | Improved |
| add COF on top of RFM (ablation) | Ours-RF · FGVC-Aircraft | wAP: 94.0 -> 94.7 | Improved |
| add RFM module to the baseline (ablation) | Baseline · Stanford Cars | wAP: 93.6 -> 94.2 | Improved |
| add COF on top of RFM (ablation) | Ours-RF · Stanford Cars | wAP: 94.2 -> 95.3 | Improved |
| unfreeze the pretrained trunk f (ablation) | Baseline · CUB (4 levels) | wAP: 88.9 -> 88.8 | No clear effect |
Closest papers by meaning
- Learned Region Sparsity and Diversity Also Predicts Visual Attention (NeurIPS 2016 · similarity 0.87)
- Which and How Many Regions to Gaze: Focus Discriminative Regions for Fine-Grained Visual Categorization (IJCV 2019 · similarity 0.87)
- Learning From Human Attention for Attribute-Assisted Visual Recognition (TPAMI 2024 · similarity 0.86)
- Dual Cross-Attention Learning for Fine-Grained Visual Categorization and Object Re-Identification (CVPR 2022 · similarity 0.86)
- Improving Fine-Grained Visual Recognition in Low Data Regimes via Self-Boosting Attention Mechanism (ECCV 2022 · similarity 0.86)
- Hierarchical Self-Attention: Generalizing Neural Attention Mechanics to Multi-Scale Problems (NeurIPS 2025 · similarity 0.85)
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.
