基于全卷积神经网络的多尺度人脸检测

Multi-Scale Face Detection Based on Full Convolution Neural Network

如何快速而准确地定位到人脸,针对这个问题,提出了一种基于全卷积神经网络的多尺度人脸检测方法.首先用全卷积层替换VGG网络中的全连接层,然后用二分类代替分类层,最后进行该算法下的人脸检测,通过对待检测的图片进行多尺度变换并将其输入到全卷积神经网络中,得到相应的概率矩阵,人脸图框通过非极大值抑制法获取.试验结果表明,该算法的准确率较高,检测时间短,性能较好.

In the report,aimed at detecting faces quickly and accurately,a multi-scale face detection method based on full Convolutional Neural Network(CNN)is proposed.Firstly,the full connectivity layer of the convolutional neural network model VGG is changed to full convolution layer.Secondly,the layer is divided into two categories of face and non-face.Finally,when the trained classification model is used for face detection,the image to be detected is input to the full convolutional network through multi-scale transformation to obtain the probability characteristic figure,and the most accurate face frame is obtained by the inhibition of non-maximal value.The experimental results show that the proposed algorithm has high detection accuracy,short detection time and good performance in face detection.