Seeing Small Faces from Robust Anchor's Perspective

02/25/2018
by   Chenchen Zhu, et al.
0

This paper introduces a novel anchor design to support anchor-based face detection for superior scale-invariant performance, especially on tiny faces. To achieve this, we explicitly address the problem that anchor-based detectors drop performance drastically on faces with tiny sizes, e.g. less than 16x16 pixels. In this paper, we investigate why this is the case. We discover that current anchor design cannot guarantee high overlaps between tiny faces and anchor boxes, which increases the difficulty of training. The new Expected Max Overlapping (EMO) score is proposed which can theoretically explain the low overlapping issue and inspire several effective strategies of new anchor design leading to higher face overlaps, including anchor stride reduction with new network architectures, extra shifted anchors, and stochastic face shifting. Comprehensive experiments show that our proposed method significantly outperforms the baseline anchor-based detector, while consistently achieving state-of-the-art results on challenging face detection datasets with competitive runtime speed.

READ FULL TEXT

page 1

page 3

page 5

page 13

page 14

research
12/20/2018

SFA: Small Faces Attention Face Detector

In recent year, tremendous strides have been made in face detection than...
research
12/19/2019

HAMBox: Delving into Online High-quality Anchors Mining for Detecting Outer Faces

Current face detectors utilize anchors to frame a multi-task learning pr...
research
03/10/2021

Wide Aspect Ratio Matching for Robust Face Detection

Recently, anchor-based methods have achieved great progress in face dete...
research
01/16/2019

DAFE-FD: Density Aware Feature Enrichment for Face Detection

Recent research on face detection, which is focused primarily on improvi...
research
05/09/2018

Anchor Cascade for Efficient Face Detection

Face detection is essential to facial analysis tasks such as facial reen...
research
04/18/2018

SFace: An Efficient Network for Face Detection in Large Scale Variations

Face detection serves as a fundamental research topic for many applicati...
research
12/13/2018

FA-RPN: Floating Region Proposals for Face Detection

We propose a novel approach for generating region proposals for performi...

Please sign up or login with your details

Forgot password? Click here to reset