Do we really need to collect millions of faces for effective face recognition?

Abstract

Face recognition capabilities have recently made extraordi- nary leaps. Though this progress is at least partially due to ballooning training set sizes - huge numbers of face images downloaded and labeled for identity - it is not clear if the formidable task of collecting so many images is truly necessary. We propose a far more accessible means of increasing training data sizes for face recognition systems: Domain spe- cific data augmentation. We describe novel methods of enriching an exist- ing dataset with important facial appearance variations by manipulating the faces it contains. This synthesis is also used when matching query images represented by standard convolutional neural networks. The effect of training and testing with synthesized images is tested on the LFW and IJB-A (verification and identification) benchmarks and Janus CS2. The performances obtained by our approach match state of the art results reported by systems trained on millions of downloaded images.