Resolving Camera Position for a Practical Application of Gaze Estimation on Edge Devices

TL;DR

This paper investigates optimal camera positioning for gaze estimation in practical edge device scenarios, demonstrating effective setup and real-time performance on NVIDIA Jetson TX2 with minimal accuracy loss.

Contribution

It introduces a logical camera setup method for gaze estimation and applies few-shot learning to achieve real-time inference on edge devices.

Findings

Achieved 12 FPS on NVIDIA Jetson TX2

Optimal camera positioning maintains accuracy

Reduced training samples with few-shot learning

Abstract

Most Gaze estimation research only works on a setup condition that a camera perfectly captures eyes gaze. They have not literarily specified how to set up a camera correctly for a given position of a person. In this paper, we carry out a study on gaze estimation with a logical camera setup position. We further bring our research in a practical application by using inexpensive edge devices with a realistic scenario. That is, we first set up a shopping environment where we want to grasp customers gazing behaviors. This setup needs an optimal camera position in order to maintain estimation accuracy from existing gaze estimation research. We then apply the state-of-the-art of few-shot learning gaze estimation to reduce training sampling in the inference phase. In the experiment, we perform our implemented research on NVIDIA Jetson TX2 and achieve a reasonable speed, 12 FPS which is faster compared with our reference work, without much degradation of gaze estimation accuracy. The source code is released at https://github.com/linh-gist/GazeEstimationTX2.