Dual-Path Enhancements in Event-Based Eye Tracking: Augmented Robustness and Adaptive Temporal Modeling

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Truong, Hoang M., Ly, Vinh-Thuan, Tran, Huy G., Nguyen, Thuan-Phat, Doan, Tram T.
Formato: Preprint
Publicado: 2025
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866908382547410944
author Truong, Hoang M.
Ly, Vinh-Thuan
Tran, Huy G.
Nguyen, Thuan-Phat
Doan, Tram T.
author_facet Truong, Hoang M.
Ly, Vinh-Thuan
Tran, Huy G.
Nguyen, Thuan-Phat
Doan, Tram T.
contents Event-based eye tracking has become a pivotal technology for augmented reality and human-computer interaction. Yet, existing methods struggle with real-world challenges such as abrupt eye movements and environmental noise. Building on the efficiency of the Lightweight Spatiotemporal Network-a causal architecture optimized for edge devices-we introduce two key advancements. First, a robust data augmentation pipeline incorporating temporal shift, spatial flip, and event deletion improves model resilience, reducing Euclidean distance error by 12% (1.61 vs. 1.70 baseline) on challenging samples. Second, we propose KnightPupil, a hybrid architecture combining an EfficientNet-B3 backbone for spatial feature extraction, a bidirectional GRU for contextual temporal modeling, and a Linear Time-Varying State-Space Module to adapt to sparse inputs and noise dynamically. Evaluated on the 3ET+ benchmark, our framework achieved 1.61 Euclidean distance on the private test set of the Event-based Eye Tracking Challenge at CVPR 2025, demonstrating its effectiveness for practical deployment in AR/VR systems while providing a foundation for future innovations in neuromorphic vision.
format Preprint
id arxiv_https___arxiv_org_abs_2504_09960
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Dual-Path Enhancements in Event-Based Eye Tracking: Augmented Robustness and Adaptive Temporal Modeling
Truong, Hoang M.
Ly, Vinh-Thuan
Tran, Huy G.
Nguyen, Thuan-Phat
Doan, Tram T.
Computer Vision and Pattern Recognition
Event-based eye tracking has become a pivotal technology for augmented reality and human-computer interaction. Yet, existing methods struggle with real-world challenges such as abrupt eye movements and environmental noise. Building on the efficiency of the Lightweight Spatiotemporal Network-a causal architecture optimized for edge devices-we introduce two key advancements. First, a robust data augmentation pipeline incorporating temporal shift, spatial flip, and event deletion improves model resilience, reducing Euclidean distance error by 12% (1.61 vs. 1.70 baseline) on challenging samples. Second, we propose KnightPupil, a hybrid architecture combining an EfficientNet-B3 backbone for spatial feature extraction, a bidirectional GRU for contextual temporal modeling, and a Linear Time-Varying State-Space Module to adapt to sparse inputs and noise dynamically. Evaluated on the 3ET+ benchmark, our framework achieved 1.61 Euclidean distance on the private test set of the Event-based Eye Tracking Challenge at CVPR 2025, demonstrating its effectiveness for practical deployment in AR/VR systems while providing a foundation for future innovations in neuromorphic vision.
title Dual-Path Enhancements in Event-Based Eye Tracking: Augmented Robustness and Adaptive Temporal Modeling
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2504.09960