AGVBench: A Reliability-Oriented Benchmark of Data Augmentation for Vein Recognition
Vein recognition is a secure biometric technology often constrained by limited annotated data and imaging variations. While data augmentation mitigates this, strategies designed for natural images may disrupt the fine-grained topology and textures essential for identity discrimination. We present AGVBench, which evaluates 30 representative augmentation strategies on five public palm- and finger-vein datasets with seven backbone architectures, covering classic CNNs, vision transformers, and vein-specific recognition models. Our results show that multi-image mixing methods (e.g., MixUp, PuzzleMi
Lineage graph
Paper → model → repo connections mined from source citations (Tier-1 exact match).
Why these links exist
- Linked via arxiv authorHaiyang Li →
AGVBench: A Reliability-Oriented Benchmark of Data Augmentation for Vein Recognition
- Linked via arxiv authorYuming Fu →
AGVBench: A Reliability-Oriented Benchmark of Data Augmentation for Vein Recognition
- Linked via arxiv authorQun Song →
AGVBench: A Reliability-Oriented Benchmark of Data Augmentation for Vein Recognition
- Linked via arxiv authorHongchao Liao →
AGVBench: A Reliability-Oriented Benchmark of Data Augmentation for Vein Recognition
- Linked via arxiv authorJing Chen →
AGVBench: A Reliability-Oriented Benchmark of Data Augmentation for Vein Recognition
- Linked via arxiv authorMounim A. EI-Yacoubi →
AGVBench: A Reliability-Oriented Benchmark of Data Augmentation for Vein Recognition
- Linked via arxiv authorXin Jin →
AGVBench: A Reliability-Oriented Benchmark of Data Augmentation for Vein Recognition