Face recognition in the wild is now advancing towards light-weight models, fast inference speed and resolution-adapted capability. In this paper, we propose a bridge distillation approach to turn a complex face model pretrained on private high-resolution faces into a light-weight one for low-resolution face recognition. In our approach, such a cross-dataset resolution-adapted knowledge transfer problem is solved via two-step distillation. In the first step, we conduct cross-dataset distillation to transfer the prior knowledge from private high-resolution faces to public high-resolution faces and generate compact and discriminative features. In the second step, the resolution-adapted distillation is conducted to further transfer the prior knowledge to synthetic low-resolution faces via multi-task learning. By learning low-resolution face representations and mimicking the adapted high-resolution knowledge, a light-weight student model can be constructed with high efficiency and promising accuracy in recognizing low-resolution faces. Experimental results show that the student model performs impressively in recognizing low-resolution faces with only 0.21M parameters and 0.057MB memory. Meanwhile, its speed reaches up to 14,705, 934 and 763 faces per second on GPU, CPU and mobile phone, respectively.

中文翻译：

---

通过桥蒸馏进行的低分辨率人脸识别

野外人脸识别现在正朝着轻量级模型，快速推理速度和适应分辨率的方向发展。在本文中，我们提出了一种桥梁蒸馏方法，将预先在私人高分辨率面部上训练的复杂面部模型转换为用于低分辨率面部识别的轻量级模型。在我们的方法中，通过两步蒸馏解决了这种跨数据集分辨率适应的知识转移问题。第一步，我们进行跨数据集蒸馏，以将先验知识从私人高分辨率面孔转移到公共高分辨率面孔，并生成紧凑而有区别的特征。在第二步中，进行适应分辨率的蒸馏，以通过多任务学习将先验知识进一步转移到合成的低分辨率面部。通过学习低分辨率人脸表示并模仿适应的高分辨率知识，可以在识别低分辨率人脸时以高效率和有希望的准确性构建轻量级学生模型。实验结果表明，该学生模型在仅使用0.21M参数和0.057MB内存的低分辨率人脸识别中表现出色。同时，在GPU，CPU和移动电话上，其速度分别达到每秒14705、934和763个面。