Research map / Label-efficient and robust learning / Recognition and long-tail
Your "Flamingo" is My "Bird": Fine-Grained, or Not
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 |
|---|---|---|---|
| add fine-grained features to coarse-label classifiers while restricting gradient flow to each level's own features (ablation) | Ours_single · CUB-200-2011 | avg_acc: 87.50 -> 88.24 | Helped |
| add fine-grained features to coarse-label classifiers while restricting gradient flow to each level's own features (ablation) | Ours_single · FGVC-Aircraft | avg_acc: 89.36 -> 90.71 | Helped |
| add fine-grained features to coarse-label classifiers while restricting gradient flow to each level's own features (ablation) | Ours_single · Stanford Cars | avg_acc: 92.18 -> 92.62 | Helped |
| split shared features into granularity-specific classifier features (ablation) | Vanilla_single · CUB-200-2011 | avg_acc: 85.77 -> 87.50 | Helped |
| split shared features into granularity-specific classifier features (ablation) | Vanilla_single · FGVC-Aircraft | avg_acc: 88.60 -> 89.36 | Helped |
| split shared features into granularity-specific classifier features (ablation) | Vanilla_single · Stanford Cars | avg_acc: 91.98 -> 92.18 | No clear effect |
| replace the manual semantic hierarchy with Hierarchical Clustering (ablation) | Ours · CUB-200-2011 | accuracy: 86.8 -> 85.0 | Hurt |
| replace the manual semantic hierarchy with Hierarchical Clustering (ablation) | Ours · FGVC-Aircraft | accuracy: 92.8 -> 90.6 | Hurt |
| replace the manual semantic hierarchy with Hierarchical Clustering (ablation) | Ours · Stanford Cars | accuracy: 94.3 -> 92.8 | Hurt |
| replace the manual semantic hierarchy with Deep Fuzzy Tree (ablation) | Ours · CUB-200-2011 | accuracy: 86.8 -> 85.5 | Hurt |
| replace the manual semantic hierarchy with Deep Fuzzy Tree (ablation) | Ours · FGVC-Aircraft | accuracy: 92.8 -> 91.7 | Hurt |
| replace the manual semantic hierarchy with Deep Fuzzy Tree (ablation) | Ours · Stanford Cars | accuracy: 94.3 -> 93.2 | Hurt |
Closest papers by meaning
- Fine-grained Classes and How to Find Them (ICML 2024 · similarity 0.89)
- Democratizing Fine-grained Visual Recognition with Large Language Models (ICLR 2024 · similarity 0.86)
- CleverBirds: A Multiple-Choice Benchmark for Fine-grained Human Knowledge Tracing (NeurIPS 2025 · similarity 0.85)
- Advancing Fine-Grained Classification by Structure and Subject Preserving Augmentation (NeurIPS 2024 · similarity 0.85)
- Multi-scale Activation, Refinement, and Aggregation: Exploring Diverse Cues for Fine-Grained Bird Recognition (AAAI 2025 · similarity 0.84)
- Novel Class Discovery for Ultra-Fine-Grained Visual Categorization (CVPR 2024 · 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.