Event-based Extraction of Navigation Features from Unsupervised Learning of Optic Flow Patterns

Abstract

We developed a Spiking Neural Network composed of two layers that processes event-based data captured by a dynamic vision sensor during navigation conditions. The training of the network was performed using a biologically plausible and unsupervised learning rule, Spike-Timing-Dependent Plasticity. With such an approach, neurons in the network naturally become selective to different components of optic flow, and a simple classifier is able to predict self-motion properties from the neural population output spiking activity. Our network has a simple architecture and a restricted number of neurons. Therefore, it is easy to implement on a neuromorphic chip and could be used for embedded applications necessitating low energy consumption.

Event Cameras in Automotive Sensing: A Review

“…-Fricker et al [115] explore the unsupervised learning of optic flow patterns for navigation features extraction, indicating potential for real-time automotive applications. dling sparse data, adapting to high-speed dynamics, coping with data variability, creating accurate annotations, managing model complexity for real-time performance, integrating data from various sensors, limited benchmark datasets, adhering to industry regulations, ensuring real-time processing, and validating system performance in real-world settings.…”