【摘要】 利用残差网络ResNet50来提取特征,基于通用目标检测领域内先进的模型Faster R-CNN作为基本模型,考虑到人脸的周围环境问题,结合注意力机制,设计一个精度达到领先水平的人脸检测模型,模型的检测速度也达到3FPS左右。利用一个在ImageNet上面预训练的ResNet50模型作为基本模型,在WIDER FACE数据集上进行训练和测试。更多还原

【Abstract】 Based on the advanced model Faster R-CNN in the field of general object detection, designs a face detection model with leading accuracy by using the residual network ResNet50 to extract features. Considering the surrounding environment of the face and the attention mecha?nism, the detection speed of the model is about 3 FPS. A ResNet50 model pre-trained on ImageNet is used as the basic model to train and test on WIDER FACE data set.更多还原