Large Pose Face Recognition via Facial Representation Learning

Overcoming image acquisition perspectives and face pose variations is a key problem in unconstrained face recognition tasks. One of the practical approaches is by reconstructing the face with extreme pose into a version that is more easily recognized by the discriminator, such as a frontal face. Often, existing methods attempt to balance the accuracy of downstream tasks with human visual perception, but ignore the differences in propensity between the two. Besides, large-scale datasets of profile-frontal paired face images are absent, which further hinders the training of models. In this work, we investigate a variety of face reconstruction approaches and propose a very simple, but very effective method to match face images across different scenes, named facial representation learning (FRL). The core idea of FRL is to introduce a representation generator in front of a pre-trained face recognition model, which can extract face representations from arbitrary faces that are more suitable for recognition model discrimination. In particular, the representation generator reconstructs the facial representation by minimising identity differences from the frontal face and adds pixel-level and adversarial constraints to cater for discriminator preferences. Extensive benchmark experiments show that the proposed method not only achieves better performance than state-of-the-art methods, but also can further squeeze the inference potential of existing face recognition models.

I. Introduction

Face recognition has been one of the most widely researched topics in computer vision for decades. With the rise of deep learning techniques and the growth in the size of internet data, face recognition methods achieve excellent performance in ’in the wild’ conditions. Supported by better depth models [1], [2], [3], more accurate loss constraints [4], [5], [6] and larger scale training data [7], [8], [9], current face recognition methods almost outperform humans on major benchmarks. However, in some extreme cases, this remains a challenging task. To date, how to effectively overcome image acquisition perspective and face pose variation is still a hot research topic, and it is the key bottleneck that has the greatest impact on face recognition performance in many real-world scenarios.