Adaptive random down-sampling data augmentation and area attention pooling for low resolution face recognition

作     者：Gao, Xuliang Sun, Yubin Xiao, Yao Gu, Yun Chai, Shuiqin Chen, Bin 

作者机构：Southwest Univ Coll Elect & Informat Engn Chongqing Key Lab Nonlinear Circuit & Intelligent Chongqing 400715 Peoples R China Chongqing Univ Sci & Technol Coll Chem & Chem Engn 20 East Daxuecheng Rd Chongqing 401331 Peoples R China 

出 版 物：《EXPERT SYSTEMS WITH APPLICATIONS》 (专家系统及其应用)

年 卷 期：2022年第209卷

核心收录：

学科分类：1201[管理学-管理科学与工程(可授管理学、工学学位)] 0808[工学-电气工程] 08[工学] 0812[工学-计算机科学与技术（可授工学、理学学位）] 

基　　金：National Nature Science Foundation of China Chongqing University of Science and Technology [ckrc2019031] 

主　　题：Low resolution face recognition Convolutional neural networks Feature extraction Pooling Data augmentation Contrastive learning 

摘      要：State-of-the-art face recognition method have achieved satisfactory performance on various face datasets, but limited to high-resolution (HR) faces. In real scenes, with affects from camera quality or distance etc., low quality face images are often obtained in surveillance scenes, which leads to poor performance of general face recognition approaches. In this work, we explored the reasons why low-resolution (LR) images cannot be coped with the conventional models, and proposed the optimization methods. The main contribution of this work includes two aspects: (1) We proposed an adaptive random down-sampling data augmentation method with the corresponding loss function and training strategy. For the network training, we mapped the face images with different resolutions to the same embedding space, and then explicitly separated different classes of LR and gathered the same class of HR and LR. (2) A pooling module based on attention mechanism, area attention pooling, was proposed. After pooling, the fine-grained information of the identity is retained, and the redundant information of the non-identity is removed. We conducted a series of visualization experiments to prove the effectiveness of our method. We evaluated our methods on the real-world LR face dataset Surveillance Cameras Face Database (SCface) and the public face evaluation dataset Labeled Faces in the Wild (LFW). Compared with state-of-the-art methods, the proposed method achieves satisfactory performance at LR images on SCface and LFW dataset.