2018 IEEE Winter Conference on Applications of Computer Vision (WACV) (2018)
Lake Tahoe, NV, USA
Mar 12, 2018 to Mar 15, 2018
In this paper, we introduce the Face Magnifier Network (Face-MageNet), a face detector based on the Faster-RCNN framework which enables the flow of discriminative information of small scale faces to the classifier without any skip or residual connections. To achieve this, Face-MagNet deploys a set of ConvTranspose, also known as deconvolution, layers in the Region Proposal Network (RPN) and another set before the Region of Interest (RoI) pooling layer to facilitate detection of finer faces. In addition, we also design, train, and evaluate three other well-tuned architectures that represent the conventional solutions to the scale problem: context pooling, skip connections, and scale partitioning. Each of these three networks achieves comparable results to the state-of-the-art face detectors. With extensive experiments, we show that Face-MagNet based on a VGG16 architecture achieves better results than the recently proposed ResNet101-based HR  method on the task of face detection on WIDER  dataset and also achieves similar results on the hard set as our other method SSH .
convolution, deconvolution, face recognition, feedforward neural nets, image classification, object detection, recurrent neural nets
P. Samangouei, R. Chellappa, M. Najibi and L. S. Davis, "Face-MagNet: Magnifying Feature Maps to Detect Small Faces," 2018 IEEE Winter Conference on Applications of Computer Vision (WACV), Lake Tahoe, NV, USA, 2018, pp. 122-130.