CrossGaze: A Strong Method for 3D Gaze Estimation in the Wild

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Cătrună, Andy, Cosma, Adrian, Rădoi, Emilian
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916123375566848
author Cătrună, Andy
Cosma, Adrian
Rădoi, Emilian
author_facet Cătrună, Andy
Cosma, Adrian
Rădoi, Emilian
contents Gaze estimation, the task of predicting where an individual is looking, is a critical task with direct applications in areas such as human-computer interaction and virtual reality. Estimating the direction of looking in unconstrained environments is difficult, due to the many factors that can obscure the face and eye regions. In this work we propose CrossGaze, a strong baseline for gaze estimation, that leverages recent developments in computer vision architectures and attention-based modules. Unlike previous approaches, our method does not require a specialised architecture, utilizing already established models that we integrate in our architecture and adapt for the task of 3D gaze estimation. This approach allows for seamless updates to the architecture as any module can be replaced with more powerful feature extractors. On the Gaze360 benchmark, our model surpasses several state-of-the-art methods, achieving a mean angular error of 9.94 degrees. Our proposed model serves as a strong foundation for future research and development in gaze estimation, paving the way for practical and accurate gaze prediction in real-world scenarios.
format Preprint
id arxiv_https___arxiv_org_abs_2402_08316
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle CrossGaze: A Strong Method for 3D Gaze Estimation in the Wild
Cătrună, Andy
Cosma, Adrian
Rădoi, Emilian
Computer Vision and Pattern Recognition
Gaze estimation, the task of predicting where an individual is looking, is a critical task with direct applications in areas such as human-computer interaction and virtual reality. Estimating the direction of looking in unconstrained environments is difficult, due to the many factors that can obscure the face and eye regions. In this work we propose CrossGaze, a strong baseline for gaze estimation, that leverages recent developments in computer vision architectures and attention-based modules. Unlike previous approaches, our method does not require a specialised architecture, utilizing already established models that we integrate in our architecture and adapt for the task of 3D gaze estimation. This approach allows for seamless updates to the architecture as any module can be replaced with more powerful feature extractors. On the Gaze360 benchmark, our model surpasses several state-of-the-art methods, achieving a mean angular error of 9.94 degrees. Our proposed model serves as a strong foundation for future research and development in gaze estimation, paving the way for practical and accurate gaze prediction in real-world scenarios.
title CrossGaze: A Strong Method for 3D Gaze Estimation in the Wild
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2402.08316