we propose a method based on deep learning, called Deep Dense Face Detector (DDFD), that does not require pose/landmark annotation and is able to detect faces in a wide range of orientations using a single model.
1. leverage the high capacity of deep convolutional networks for classification and feature extraction to learn a single classifier for detecing faces from multiple views
2. minimize the computational complexity by simplifying the architecture of the detector
The classfier consists of 8 layers where first 5 layers are convolutional and the last 3 layers are fully-connected.
First, converted the fully-connected layers into convolutional layers by reshaping layer
parameters, to efficiently run the CNN on images of any size and obtain a heat-map of the face classifier.
Each point in the heat-map shows the CNN response, the probability of having a face, for its corresponding 227×227 region in the original image.
意思是heat-map种每一个点表示的是一个概率 —— 再原图中以它为中心的227×227的区域有一张脸的概率。
Second, The detected regions were then processed by non-maximal suppression to accurately localize the faces.
Finally, we scaled the images up/down and obtained new heat-maps to detect faces of different sizes. We tried different scaling schemes and found that rescaling image 3 times per octave gives reasonably good performance.
To summarize, the proposed face detector based on deep CNN is able to detect faces from different angles and handle occlusion to some extent.
there seems to be a correlation between distribution of positive examples in the training set and scores of the proposed detector.
In future we are planning to use better sampling strategies and more sophisticated data augmentation techniques to further improve performance of the proposed method for detecting
occluded and rotated faces.