Face detection is a classical problem in computer vision. It is still a difficult task due to many nuisances that naturally occur in the wild, including extreme pose, exaggerated expressions, significant illumination variations and severe occlusion. In this paper, we propose a multi-scale fully convolutional network (MS-FCN) for face detection. To reduce computation, the intermediate convolutional feature maps (conv) are shared by every scale model. We up-sample and down-sample the final conv map to approximate K levels of a feature pyramid, leading to a wide range of face scales that can be detected. At each feature pyramid level, a FCN is trained end-to-end to deal with faces in a small range of scale change. Because of the up-sampling, our method can detect very small faces (10 × 10 pixels). We test our MS-FCN detector on four public face detection bench-marks, including FDDB, WIDER FACE, AFW and PASCAL FACE. Extensive experiments show that our detector out-performs state-of-the-art methods on all these datasets in general and by a substantial margin on the most challenging among them (e.g. WIDER FACE Hard). Also, MS-FCN runs at 23 FPS on a GPU for images of size 640 × 480 with no assumption on the minimum detectable face size.
展开▼