Group Sampling for Scale Invariant Face Detection

Detectors based on deep learning tend to detect multi-scale faces on a single input image for efficiency. Recent works, such as FPN and SSD, generally use feature maps from multiple layers with different spatial resolutions to detect objects at different scales, e.g., high-resolution feature maps for small objects. However, we find that objects at all scales can also be well detected with features from a single layer of the network. In this paper, we carefully examine the factors affecting detection performance across a large range of scales, and conclude that the balance of training samples, including both positive and negative ones, at different scales is the key. We propose a group sampling method which divides the anchors into several groups according to the scale, and ensure that the number of samples for each group is the same during training. Our approach using only one single layer of FPN is able to advance the state-of-the-arts. Comprehensive analysis and extensive experiments have been conducted to show the effectiveness of the proposed method. Moreover, we show that our approach is favorably applicable to other detection tasks and to other pipelines. Our approach achieves state-of-the-art results on several face detection benchmarks without bells and whistles.