Targeting Ultimate Accuracy: Face Recognition via Deep Embedding
This addresses face recognition accuracy for security and identification applications, representing a strong incremental improvement.
The paper tackles face recognition by proposing a two-stage approach combining multi-patch deep CNN and deep metric learning to extract discriminative features, achieving 99.77% pair-wise verification accuracy on LFW and better results under practical protocols.
Face Recognition has been studied for many decades. As opposed to traditional hand-crafted features such as LBP and HOG, much more sophisticated features can be learned automatically by deep learning methods in a data-driven way. In this paper, we propose a two-stage approach that combines a multi-patch deep CNN and deep metric learning, which extracts low dimensional but very discriminative features for face verification and recognition. Experiments show that this method outperforms other state-of-the-art methods on LFW dataset, achieving 99.77% pair-wise verification accuracy and significantly better accuracy under other two more practical protocols. This paper also discusses the importance of data size and the number of patches, showing a clear path to practical high-performance face recognition systems in real world.