Supervised Transformer Network for Efficient Face Detection
This addresses face detection in real-world scenarios with pose variations, but it is incremental as it builds on existing cascaded and RPN/RCNN methods.
The paper tackles the challenge of large pose variations in face detection by proposing a cascaded Convolutional Neural Network with a supervised transformer stage to warp candidate regions using facial landmarks, achieving state-of-the-art detection accuracies on public benchmarks and running at 30 FPS on a single CPU core for VGA-resolution images.
Large pose variations remain to be a challenge that confronts real-word face detection. We propose a new cascaded Convolutional Neural Network, dubbed the name Supervised Transformer Network, to address this challenge. The first stage is a multi-task Region Proposal Network (RPN), which simultaneously predicts candidate face regions along with associated facial landmarks. The candidate regions are then warped by mapping the detected facial landmarks to their canonical positions to better normalize the face patterns. The second stage, which is a RCNN, then verifies if the warped candidate regions are valid faces or not. We conduct end-to-end learning of the cascaded network, including optimizing the canonical positions of the facial landmarks. This supervised learning of the transformations automatically selects the best scale to differentiate face/non-face patterns. By combining feature maps from both stages of the network, we achieve state-of-the-art detection accuracies on several public benchmarks. For real-time performance, we run the cascaded network only on regions of interests produced from a boosting cascade face detector. Our detector runs at 30 FPS on a single CPU core for a VGA-resolution image.