论文标题
群集和骨料：大探头集的面部识别
Cluster and Aggregate: Face Recognition with Large Probe Set
论文作者
论文摘要
特征融合在不受约束的面部识别中起着至关重要的作用，其中输入（探针）包括一组$ n $低质量的图像，其个人品质各不相同。注意力和复发模块的进步导致了融合，可以模拟输入集中图像之间的关系。但是，由于其二次复杂性和经常性模块具有输入顺序敏感性，注意力机制无法扩展到大$ n $。我们提出了一个两阶段的特征融合范式，群集和骨料，可以扩展到大$ n $，并保持使用顺序不变性执行顺序推断的能力。具体而言，集群阶段是$ n $输入到$ m $ $ $全球集群中心的线性分配，而聚合阶段是$ m $ cluster的功能的融合。当输入是顺序的，因为它们可以用作过去特征的摘要，因此群集功能起着不可或缺的作用。通过利用增量平均操作的订单不变性，我们设计了一个实现批处订单不变性的更新规则，该规则可以保证，随着时间步长的增加，序列中早期图像的贡献不会减小。 IJB-B和IJB-S基准数据集的实验表明，在不受约束的面部识别中提出的两阶段范式的优越性。 https://github.com/mk-minchul/caface中有代码和预估计的型号
Feature fusion plays a crucial role in unconstrained face recognition where inputs (probes) comprise of a set of $N$ low quality images whose individual qualities vary. Advances in attention and recurrent modules have led to feature fusion that can model the relationship among the images in the input set. However, attention mechanisms cannot scale to large $N$ due to their quadratic complexity and recurrent modules suffer from input order sensitivity. We propose a two-stage feature fusion paradigm, Cluster and Aggregate, that can both scale to large $N$ and maintain the ability to perform sequential inference with order invariance. Specifically, Cluster stage is a linear assignment of $N$ inputs to $M$ global cluster centers, and Aggregation stage is a fusion over $M$ clustered features. The clustered features play an integral role when the inputs are sequential as they can serve as a summarization of past features. By leveraging the order-invariance of incremental averaging operation, we design an update rule that achieves batch-order invariance, which guarantees that the contributions of early image in the sequence do not diminish as time steps increase. Experiments on IJB-B and IJB-S benchmark datasets show the superiority of the proposed two-stage paradigm in unconstrained face recognition. Code and pretrained models are available in https://github.com/mk-minchul/caface
