Scale-Aware Face Detection

Zekun Hao, Yu Liu, Hongwei Qin, Junjie Yan, Xiu Li, Xiaolin Hu; The IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2017, pp. 6186-6195


Convolutional neural network (CNN) based face detectors are inefficient in handling faces of diverse scales. They rely on either fitting a large single model to faces across a large scale range or multi-scale testing. Both are computationally expensive. We propose Scale-aware Face Detection (SAFD) to handle scale explicitly using CNN, and achieve better performance with less computation cost. Prior to detection, an efficient CNN predicts the scale distribution histogram of the faces. Then the scale histogram guides the zoom-in and zoom-out of the image. Since the faces will be approximately in uniform scale after zoom, they can be detected accurately even with much smaller CNN. Actually, more than 99% of the faces in AFW can be covered with less than two zooms per image. Extensive experiments on FDDB, MALF and AFW show advantages of SAFD.

Related Material

[pdf] [arXiv]
author = {Hao, Zekun and Liu, Yu and Qin, Hongwei and Yan, Junjie and Li, Xiu and Hu, Xiaolin},
title = {Scale-Aware Face Detection},
booktitle = {The IEEE Conference on Computer Vision and Pattern Recognition (CVPR)},
month = {July},
year = {2017}