Face Detection on Mobile: Five Implementations and Analysis

TL;DR

This paper compares five face detection algorithms on mobile devices, analyzing their speed and efficiency to guide optimal selection for various mobile applications.

Contribution

It adapts and evaluates five face detection algorithms for mobile devices, providing insights into their performance and suitability for real-world applications.

Findings

Cascaded algorithms are faster in scenes without faces.

BlazeFace is slower on empty scenes.

Guidance on algorithm choice for mobile face detection.

Abstract

In many practical cases face detection on smartphones or other highly portable devices is a necessity. Applications include mobile face access control systems, driver status tracking, emotion recognition, etc. Mobile devices have limited processing power and should have long-enough battery life even with face detection application running. Thus, striking the right balance between algorithm quality and complexity is crucial. In this work we adapt 5 algorithms to mobile. These algorithms are based on handcrafted or neural-network-based features and include: Viola-Jones (Haar cascade), LBP, HOG, MTCNN, BlazeFace. We analyze inference time of these algorithms on different devices with different input image resolutions. We provide guidance, which algorithms are the best fit for mobile face access control systems and potentially other mobile applications. Interestingly, we note that cascaded algorithms perform faster on scenes without faces, while BlazeFace is slower on empty scenes. Exploiting this behavior might be useful in practice.