Beyond Trade-off: Accelerate FCN-based Face Detector with Higher Accuracy

by   Guanglu Song, et al.

Fully convolutional neural network (FCN) has been dominating the game of face detection task for a few years with its congenital capability of sliding-window-searching with shared kernels, which boiled down all the redundant calculation, and most recent state-of-the-art methods such as Faster-RCNN, SSD, YOLO and FPN use FCN as their backbone. So here comes one question: Can we find a universal strategy to further accelerate FCN with higher accuracy, so could accelerate all the recent FCN-based methods? To analyze this, we decompose the face searching space into two orthogonal directions, `scale' and `spatial'. Only a few coordinates in the space expanded by the two base vectors indicate foreground. So if FCN could ignore most of the other points, the searching space and false alarm should be significantly boiled down. Based on this philosophy, a novel method named scale estimation and spatial attention proposal (S^2AP) is proposed to pay attention to some specific scales and valid locations in the image pyramid. Furthermore, we adopt a masked-convolution operation based on the attention result to accelerate FCN calculation. Experiments show that FCN-based method RPN can be accelerated by about 4× with the help of S^2AP and masked-FCN and at the same time it can also achieve the state-of-the-art on FDDB, AFW and MALF face detection benchmarks as well.


Detecting Faces Using Region-based Fully Convolutional Networks

Face detection has achieved great success using the region-based methods...

Multi-Branch Fully Convolutional Network for Face Detection

Face detection is a fundamental problem in computer vision. It is still ...

A Multi-Scale Cascade Fully Convolutional Network Face Detector

Face detection is challenging as faces in images could be present at arb...

Insights into LSTM Fully Convolutional Networks for Time Series Classification

Long Short Term Memory Fully Convolutional Neural Networks (LSTM-FCN) an...

Efficient and Scalable View Generation from a Single Image using Fully Convolutional Networks

Single-image-based view generation (SIVG) is important for producing 3D ...

Focal FCN: Towards Small Object Segmentation with Limited Training Data

Small object segmentation is a common task in medical image analysis. Tr...

Frustum ConvNet: Sliding Frustums to Aggregate Local Point-Wise Features for Amodal 3D Object Detection

In this work, we propose a novel method termed Frustum ConvNet (F-ConvNe...

Please sign up or login with your details

Forgot password? Click here to reset