S$^3$FD: Single Shot Scale-invariant Face Detector

TL;DR

This paper introduces S$^3$FD, a real-time, scale-invariant face detector that excels at detecting faces of all sizes, especially small faces, using a single deep neural network with novel anchor strategies.

Contribution

The paper proposes a scale-equitable detection framework, a scale compensation strategy for small faces, and a max-out background label to reduce false positives, advancing face detection technology.

Findings

Achieves state-of-the-art results on multiple face detection benchmarks.

Operates at 36 FPS on a high-end GPU for VGA images.

Significantly improves detection of small faces across various datasets.

Abstract

This paper presents a real-time face detector, named Single Shot Scale-invariant Face Detector (S$^3$FD), which performs superiorly on various scales of faces with a single deep neural network, especially for small faces. Specifically, we try to solve the common problem that anchor-based detectors deteriorate dramatically as the objects become smaller. We make contributions in the following three aspects: 1) proposing a scale-equitable face detection framework to handle different scales of faces well. We tile anchors on a wide range of layers to ensure that all scales of faces have enough features for detection. Besides, we design anchor scales based on the effective receptive field and a proposed equal proportion interval principle; 2) improving the recall rate of small faces by a scale compensation anchor matching strategy; 3) reducing the false positive rate of small faces via a max-out background label. As a consequence, our method achieves state-of-the-art detection performance on all the common face detection benchmarks, including the AFW, PASCAL face, FDDB and WIDER FACE datasets, and can run at 36 FPS on a Nvidia Titan X (Pascal) for VGA-resolution images.