AI-Assisted Verification of Biometric Data Collection

TL;DR

This paper evaluates the performance of YOLO-based models for action recognition from video feeds on various Android devices, focusing on hardware limitations and model efficiency for mobile deployment.

Contribution

It compares YOLO architecture performance across different devices and discusses challenges in recognizing actions on limited hardware.

Findings

YOLO models can run on a range of Android devices with acceptable accuracy.

Hardware limitations significantly affect action recognition performance.

Model optimization is crucial for real-time mobile video analysis.

Abstract

Recognizing actions from a video feed is a challenging task to automate, especially so on older hardware. There are two aims for this project: one is to recognize an action from the front-facing camera on an Android phone, the other is to support as many phones and Android versions as possible. This limits us to using models that are small enough to run on mobile phones with and without GPUs, and only using the camera feed to recognize the action. In this paper we compare performance of the YOLO architecture across devices (with and without dedicated GPUs) using models trained on a custom dataset. We also discuss limitations in recognizing faces and actions from video on limited hardware.