Research map / Label-efficient and robust learning / Recognition and long-tail
Graph-Propagation Based Correlation Learning for Weakly Supervised Fine-Grained Image Classification
Zhuhui Wang, Shijie Wang, Haojie Li, Zhi Dou, Jianjun Li
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 default patches + score mechanism (ablation) | ResNet-50 baseline · CUB-200-2011 | accuracy: 84.5% -> 86.2% | Improved |
| add CGP criss-cross graph propagation (ablation) | BL+DP+Score · CUB-200-2011 | accuracy: 86.2% -> 87.7% | Improved |
| criss-cross propagation instead of same-frequency aggregation (ablation) | CGP · CUB-200-2011 | accuracy: 87.2% -> 87.7% | Improved |
| add CFS feature strengthening GCN (ablation) | BL+DP+Score+CGP · CUB-200-2011 | accuracy: 87.7% -> 88.3% | Improved |
| increase CGP depth 3 -> 4 (ablation) | GCL · CUB-200-2011 | accuracy: 87.8% -> 87.5% | Hurt |
| increase CFS depth 3 -> 4 (ablation) | GCL · CUB-200-2011 | accuracy: 88.3% -> 88.1% | Hurt |
| increase selected patches K 2 -> 4 (ablation) | GCL · CUB-200-2011 | accuracy: 87.80% -> 88.30% | Improved |
| GCL vs DCL | fine-grained classifier · Stanford Cars | accuracy: 94.5% vs 94.0% | Hurt |
Closest papers by meaning
- Weakly Supervised Fine-Grained Image Classification via Guassian Mixture Model Oriented Discriminative Learning (CVPR 2020 · similarity 0.88)
- Interweaving Insights: High-Order Feature Interaction for Fine-Grained Visual Recognition (IJCV 2024 · similarity 0.86)
- Few-Shot Fine-Grained Image Classification with Progressively Feature Refinement and Continuous Relationship Modeling (AAAI 2025 · similarity 0.85)
- Cross-Layer and Cross-Sample Feature Optimization Network for Few-Shot Fine-Grained Image Classification (AAAI 2024 · similarity 0.84)
- Novel Class Discovery for Ultra-Fine-Grained Visual Categorization (CVPR 2024 · similarity 0.84)
- HiGDA: Hierarchical Graph of Nodes to Learn Local-to-Global Topology for Semi-Supervised Domain Adaptation (AAAI 2025 · similarity 0.83)
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.
