Pooling Faces: Template based Face Recognition with Pooled Face Images
This work addresses efficiency and accuracy challenges in face recognition systems, offering a novel application of pooling that is competitive with deep learning methods.
The paper tackles template-based face recognition by using average pooling of face images partitioned by quality and pose, achieving state-of-the-art accuracy on IJB-A and Janus CS2 benchmarks while reducing computational and storage costs.
We propose a novel approach to template based face recognition. Our dual goal is to both increase recognition accuracy and reduce the computational and storage costs of template matching. To do this, we leverage on an approach which was proven effective in many other domains, but, to our knowledge, never fully explored for face images: average pooling of face photos. We show how (and why!) the space of a template's images can be partitioned and then pooled based on image quality and head pose and the effect this has on accuracy and template size. We perform extensive tests on the IJB-A and Janus CS2 template based face identification and verification benchmarks. These show that not only does our approach outperform published state of the art despite requiring far fewer cross template comparisons, but also, surprisingly, that image pooling performs on par with deep feature pooling.