SFace: An Efficient Network for Face Detection in Large Scale Variations
Jianfeng Wang12∗ Ye Yuan 1† Boxun Li† Gang Yu† Sun Jian†
Abstract
Face detection serves as a fundamental research topic for many applications like face recognition. Impressive progress has been made especially with the recent development of convolutional neural networks. However, the issue of large scale variations, which widely exists in high resolution images/videos, has not been well addressed in the literature. In this paper, we present a novel algorithm called SFace, which efficiently integrates the anchor-based method and anchor-free method to address the scale issues. A new dataset called 4K-Face is also introduced to evaluate the performance of face detection with extreme large scale variations. The SFace architecture shows promising results on the new 4K-Face benchmarks. In addition, our method can run at $\sim$ 50 frames per second (fps) with an accuracy of 80% AP on the standard WIDER FACE dataset, which outperforms the state-of-art algorithms by almost one order of magnitude in speed while achieves comparative performance.
中文速览
人脸检测在高分辨率图像（如4K视频）中面临极端尺度变化的难题——同一画面里人脸大小可能从10×10像素跨越到2000×2000像素，现有方法要么依赖精心设计的锚框（anchor）导致泛化性差，要么采用无锚框方式但定位精度不足。为此，研究者提出了SFace算法，将基于锚框的检测分支（负责中小尺度人脸的精准定位）与无锚框分支（负责捕捉超大尺度人脸）融合在同一个特征金字塔网络中，并设计了一种基于交并比（IOU）预测的重评分机制，统一两条分支的置信度，使检测结果能够无缝合并。同时，研究者还发布了首个专门针对4K超高清图像的人脸检测基准数据集4K-Face，包含约5000张图像和3万条标注。实验表明，SFace在标准WIDER FACE数据集上以约50帧/秒的速度达到80% AP，速度比当时最先进方法快近一个数量级，同时在4K-Face上也取得了领先表现，对推动高分辨率实时人脸检测的实际应用具有重要意义。
原文 arXiv:1804.06559;中英对照 + 大白话阅读 https://aha.fim.ai/paper/1804.06559v2