An event-driven framework for fly-inspired visual motion detection

2026-07-06

AI Summary

This work addresses the challenge of motion detection with event cameras under low-light conditions, where noise susceptibility and the difficulty of effectively integrating biologically inspired neural processing hinder performance. To this end, the paper proposes a training-free, event-driven motion detection framework that uniquely integrates temporal surface encoding with a fruit fly lamina-medulla feedforward neural circuit model. A bottom-up attention mechanism is incorporated to suppress background motion and enhance foreground saliency. The resulting system exhibits low latency, high energy efficiency, interpretable parameters, and robust noise resilience, enabling real-time motion direction estimation on real-world automotive datasets. It outperforms conventional frame-based models and existing optimization-based approaches, successfully combining the temporal precision of event-based vision with the computational efficiency of biological neural processing.

Abstract

Fast and reliable motion detection is essential for machine vision and autonomous systems operating in dynamic environments. This work integrates emerging event-based sensing with biologically structured neural computation to establish an efficient computational paradigm for visual motion detection. The proposed framework is built upon a recently developed fly-inspired neural network that emulates motion-processing circuits in the optic lobe. Owing to its feed-forward and training-free architecture, the neural model requires only a small number of interpretable parameters and is well suited for real-time embedded implementation. Event cameras provide low-latency, low-power, and high-dynamic-range visual sensing by asynchronously transmitting brightness-change events. However, their performance can be degraded by event noise, including temporal noise and junction-leakage-induced activity, particularly under low-light conditions. Moreover, effective integration between event-based visual representations and biologically inspired neural processing remains under-explored. To address these challenges, we propose an event-driven computational framework that combines time-surface encoding for front-end event representation with a fly optic-lobe-inspired neural network for foreground motion-direction estimation. A bottom-up attention mechanism is further incorporated to suppress background motion and enhance the saliency of foreground targets. The proposed method is evaluated on real-world ground-vehicle datasets and compared with a baseline frame-based model and an optimization-based approach. Experimental results demonstrate that the framework effectively combines the temporal advantages of event-driven vision with the efficiency and interpretability of bio-inspired neural processing.

Problem

• event-based vision
• motion detection
• neural computation
• event noise
• bio-inspired processing

Innovation

• event-based vision
• fly-inspired neural network
• time-surface encoding
• motion detection
• bio-inspired attention