Face Detection through Scale-Friendly Deep Convolutional Networks

TL;DR

This paper introduces ScaleFace, a unified deep convolutional network designed for face detection across a wide range of scales, achieving high accuracy and efficiency without the need for image pyramids.

Contribution

The paper presents a novel scale-friendly deep network that models faces of various sizes with specialized sub-networks, integrated into a single end-to-end trainable model, improving over existing methods.

Findings

Achieves 76.4% AP on WIDER FACE dataset.

Reaches 96% recall on FDDB dataset.

Operates at 7 fps for 900x1300 images.

Abstract

In this paper, we share our experience in designing a convolutional network-based face detector that could handle faces of an extremely wide range of scales. We show that faces with different scales can be modeled through a specialized set of deep convolutional networks with different structures. These detectors can be seamlessly integrated into a single unified network that can be trained end-to-end. In contrast to existing deep models that are designed for wide scale range, our network does not require an image pyramid input and the model is of modest complexity. Our network, dubbed ScaleFace, achieves promising performance on WIDER FACE and FDDB datasets with practical runtime speed. Specifically, our method achieves 76.4 average precision on the challenging WIDER FACE dataset and 96% recall rate on the FDDB dataset with 7 frames per second (fps) for 900 * 1300 input image.