HyperFace: A Deep Multi-task Learning Framework for Face Detection, Landmark Localization, Pose Estimation, and Gender Recognition
This work addresses the need for efficient and accurate multi-task face analysis in computer vision applications, representing an incremental advancement by building on existing CNN architectures like ResNet.
The paper tackled the problem of simultaneous face detection, landmark localization, pose estimation, and gender recognition by proposing HyperFace, a deep multi-task learning framework that fuses CNN layers and uses multi-task learning, resulting in significant performance improvements over competitive algorithms.
We present an algorithm for simultaneous face detection, landmarks localization, pose estimation and gender recognition using deep convolutional neural networks (CNN). The proposed method called, HyperFace, fuses the intermediate layers of a deep CNN using a separate CNN followed by a multi-task learning algorithm that operates on the fused features. It exploits the synergy among the tasks which boosts up their individual performances. Additionally, we propose two variants of HyperFace: (1) HyperFace-ResNet that builds on the ResNet-101 model and achieves significant improvement in performance, and (2) Fast-HyperFace that uses a high recall fast face detector for generating region proposals to improve the speed of the algorithm. Extensive experiments show that the proposed models are able to capture both global and local information in faces and performs significantly better than many competitive algorithms for each of these four tasks.