Event Based Eye Tracking

Event-based eye tracking uses specialized cameras to capture asynchronous visual events, offering high temporal resolution and low power consumption compared to traditional frame-based systems. Current research focuses on developing efficient and accurate algorithms, including convolutional neural networks (CNNs), recurrent neural networks (RNNs like LSTMs and GRUs), and spiking neural networks (SNNs), often incorporating novel event representations and loss functions to optimize pupil localization. This technology is crucial for applications demanding real-time performance and low latency, such as extended reality (XR) interfaces, wearable health monitoring, and biometric authentication, driving significant advancements in both hardware and software.