Face Detection on Surveillance Images

TL;DR

This paper introduces a body pose based face detection method tailored for surveillance images, demonstrating superior accuracy and lower false alarms compared to traditional methods in challenging surveillance scenarios.

Contribution

The paper proposes a novel body pose based face detection approach and provides a comprehensive performance comparison with existing methods in surveillance environments.

Findings

Our method achieves higher accuracy in surveillance scenarios.

It results in fewer false alarms compared to traditional face detection methods.

The approach has competitive detection times.

Abstract

In last few decades, a lot of progress has been made in the field of face detection. Various face detection methods have been proposed by numerous researchers working in this area. The two well-known benchmarking platform: the FDDB and WIDER face detection provide quite challenging scenarios to assess the efficacy of the detection methods. These benchmarking data sets are mostly created using images from the public network ie. the Internet. A recent, face detection and open-set recognition challenge has shown that those same face detection algorithms produce high false alarms for images taken in surveillance scenario. This shows the difficult nature of the surveillance environment. Our proposed body pose based face detection method was one of the top performers in this competition. In this paper, we perform a comparative performance analysis of some of the well known face detection methods including the few used in that competition, and, compare them to our proposed body pose based face detection method. Experiment results show that, our proposed method that leverages body information to detect faces, is the most realistic approach in terms of accuracy, false alarms and average detection time, when surveillance scenario is in consideration.