Research map / Detection and segmentation / CNN features
Filtration and Distillation: Enhancing Region Attention for Fine-Grained Visual Categorization
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 filtration learning (ablation) | ResNet50 · CUB-200-2011 | accuracy: 83.54 -> 88.07 | Helped |
| add distillation learning (ablation) | ResNet50 · CUB-200-2011 | accuracy: 83.54 -> 84.83 | Helped |
| add distillation learning on top of filtration learning (ablation) | ResNet50 + FL · CUB-200-2011 | accuracy: 88.07 -> 88.59 | Helped |
| add point-wise loss to pair-wise loss (ablation) | FDL filtration · CUB-200-2011 | accuracy: 88.14 -> 88.59 | Helped |
| add pair-wise loss to point-wise loss (ablation) | FDL filtration · CUB-200-2011 | accuracy: 88.01 -> 88.59 | Helped |
| number of attention regions K (best 4, 3 for cars) (ablation) | FDL · four FGVC datasets | accuracy | Mixed |
| FDL training for object-only inference | ResNet50 · CUB-200-2011 | accuracy: 85.49 -> 88.35 | Helped |
| FDL vs DCL | ResNet50 · Stanford Cars | accuracy: 94.5 vs 94.27 | Hurt |
Closest papers by meaning
- Attention Convolutional Binary Neural Tree for Fine-Grained Visual Categorization (CVPR 2020 · similarity 0.88)
- TransFG: A Transformer Architecture for Fine-Grained Recognition (AAAI 2022 · similarity 0.87)
- Looking for the Devil in the Details: Learning Trilinear Attention Sampling Network for Fine-Grained Image Recognition (CVPR 2019 · similarity 0.86)
- Novel Class Discovery for Ultra-Fine-Grained Visual Categorization (CVPR 2024 · similarity 0.86)
- Part-level Semantic-guided Contrastive Learning for Fine-grained Visual Classification (ICLR 2026 · similarity 0.86)
- Advancing Fine-Grained Classification by Structure and Subject Preserving Augmentation (NeurIPS 2024 · 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.