A Comparison of CNN-based Face and Head Detectors for Real-Time Video Surveillance Applications

TL;DR

This paper compares various CNN architectures for face and head detection in real-time video surveillance, evaluating their accuracy and computational efficiency to determine their practicality in real-world applications.

Contribution

It provides an empirical comparison of state-of-the-art CNN architectures for face and head detection, focusing on accuracy and real-time computational feasibility.

Findings

CNN architectures achieve high accuracy over traditional methods

Computational complexity limits real-time application viability

Region-based architectures offer a balance between accuracy and efficiency

Abstract

Detecting faces and heads appearing in video feeds are challenging tasks in real-world video surveillance applications due to variations in appearance, occlusions and complex backgrounds. Recently, several CNN architectures have been proposed to increase the accuracy of detectors, although their computational complexity can be an issue, especially for real-time applications, where faces and heads must be detected live using high-resolution cameras. This paper compares the accuracy and complexity of state-of-the-art CNN architectures that are suitable for face and head detection. Single pass and region-based architectures are reviewed and compared empirically to baseline techniques according to accuracy and to time and memory complexity on images from several challenging datasets. The viability of these architectures is analyzed with real-time video surveillance applications in mind. Results suggest that, although CNN architectures can achieve a very high level of accuracy compared to traditional detectors, their computational cost can represent a limitation for many practical real-time applications.