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

TL;DR

CrossGaze introduces a versatile, architecture-agnostic approach for 3D gaze estimation in unconstrained environments, leveraging existing models and attention modules to outperform previous methods on the Gaze360 benchmark.

Contribution

The paper presents a flexible baseline for 3D gaze estimation that integrates established computer vision architectures and attention modules without requiring specialized designs.

Findings

Achieves a mean angular error of 9.94 degrees on Gaze360

Outperforms several state-of-the-art gaze estimation methods

Provides a modular framework adaptable to future improvements

Abstract

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.