A Lightweight Spatiotemporal Network for Online Eye Tracking with Event Camera

Event-based data are commonly encountered in edge computing environments where efficiency and low latency are critical. To interface with such data and leverage their rich temporal features, we propose a causal spatiotemporal convolutional network. This solution targets efficient implementation on edge-appropriate hardware with limited resources in three ways: 1) deliberately targets a simple architecture and set of operations (convolutions, ReLU activations) 2) can be configured to perform online inference efficiently via buffering of layer outputs 3) can achieve more than 90% activation sparsity through regularization during training, enabling very significant efficiency gains on event-based processors. In addition, we propose a general affine augmentation strategy acting directly on the events, which alleviates the problem of dataset scarcity for event-based systems. We apply our model on the AIS 2024 event-based eye tracking challenge, reaching a score of 0.9916 p10 accuracy on the Kaggle private testset.

1. Introduction

Event cameras are sensors that generate outputs (events) responding to optical changes in the scene’s luminance [2]. For each pixel, an event is only produced when a brightness change is detected, which enables high levels of sparsity and temporal resolution when compared to traditional frame-based cameras. Given its high temporal resolution, often at sub-millisecond scales, data from event cameras contain very rich temporal features capturing subtle movement patterns.