CCFace: Classification Consistency for Low-Resolution Face Recognition
CCFace: Classification Consistency for Low-Resolution Face Recognition
复制标题
DOI：
10.1109/ijcb57857.2023.10448973
复制
发表时间：
2023-08
期刊：
影响因子：
--
通讯作者：
Mohammad Saeed Ebrahimi Saadabadi;Sahar Rahimi Malakshan;Hossein Kashiani;N. Nasrabadi
中科院分区：
文献类型：
--
作者：
Mohammad Saeed Ebrahimi Saadabadi;Sahar Rahimi Malakshan;Hossein Kashiani;N. Nasrabadi
近年来，深度人脸识别方法在野外数据集上表现出了令人印象深刻的结果。然而，当应用于 TinyFace 或 SCFace 等现实世界的低分辨率基准测试时，这些方法的性能显着下降。为了应对这一挑战，我们提出了一种新颖的分类一致性知识蒸馏方法，将学习的分类器从高分辨率模型转移到低分辨率网络。这种方法有助于找到低分辨率实例的区分性表示。为了进一步提高性能，我们受到流行的角度边缘损失函数成功的启发，使用自适应角度惩罚设计了知识蒸馏损失。自适应惩罚减少了低分辨率样本的过度拟合，并缓解了与数据增强集成的模型的收敛问题。此外，我们利用基于最先进的半监督表示学习范式的非对称交叉分辨率学习方法来提高低分辨率实例的辨别力并防止它们形成集群。我们提出的方法在低分辨率基准测试中的性能优于最先进的方法，在保持高分辨率基准测试性能的同时，TinyFace 的性能提高了 3%。
In recent years, deep face recognition methods have demonstrated impressive results on in-the-wild datasets. However, these methods have shown a significant decline in performance when applied to real-world low-resolution benchmarks like TinyFace or SCFace. To address this challenge, we propose a novel classification consistency knowledge distillation approach that transfers the learned classifier from a high-resolution model to a low-resolution network. This approach helps in finding discriminative representations for low-resolution instances. To further improve the performance, we designed a knowledge distillation loss using the adaptive angular penalty inspired by the success of the popular angular margin loss function. The adaptive penalty reduces overfitting on low-resolution samples and alleviates the convergence issue of the model integrated with data augmentation. Additionally, we utilize an asymmetric cross-resolution learning approach based on the state-of-the-art semi-supervised representation learning paradigm to improve discriminability on low-resolution instances and prevent them from forming a cluster. Our proposed method outperforms state-of-the-art approaches on low-resolution benchmarks, with a three percent improvement on TinyFace while maintaining performance on high-resolution benchmarks.