通过融合不同CNN模型提升远距离眼周识别精度。
Exploring Complementarity and Explainability in CNNs for Periocular Verification Across Acquisition Distances
- 训练三种复杂度模型并用逻辑回归融合得分
- 融合后在UBIPr数据集上达到新SOTA性能
- 热力图显示模型关注区域不同，解释互补性
我们在UBIPr数据集上研究了不同CNN在不同采集距离下的眼周验证互补性。在VGGFace2的大规模眼区图像上训练了三种复杂度递增的网络架构（SqueezeNet、MobileNetv2和ResNet50），采用余弦和卡方度量分析性能，比较不同初始化方式，并通过逻辑回归实现分数级融合。同时利用LIME热图与詹森-香农散度对比各模型注意力模式。虽然ResNet50单独表现最佳，但三者融合带来显著提升，尤其在全部组合时效果最优。热力图显示各网络通常聚焦于图像不同区域，解释其互补性。本方法在UBIPr上显著优于先前工作，达到新状态最优水平。
原文摘要 · Abstract (English)
We study the complementarity of different CNNs for periocular verification at different distances on the UBIPr database. We train three architectures of increasing complexity (SqueezeNet, MobileNetv2, and ResNet50) on a large set of eye crops from VGGFace2. We analyse performance with cosine and chi2 metrics, compare different network initialisations, and apply score-level fusion via logistic regression. In addition, we use LIME heatmaps and Jensen-Shannon divergence to compare attention patterns of the CNNs. While ResNet50 consistently performs best individually, the fusion provides substantial gains, especially when combining all three networks. Heatmaps show that networks usually focus on distinct regions of a given image, which explains their complementarity. Our method significantly outperforms previous works on UBIPr, achieving a new state-of-the-art.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。