Spiking Optical Flow for Event-based Sensors Using IBM's TrueNorth Neurosynaptic System

TL;DR

This paper presents a low-power, spike-based neural network system using IBM's TrueNorth for real-time optical flow estimation from event-based sensors, achieving accurate motion detection with minimal energy consumption.

Contribution

It introduces a fully spike-based neural network implementation on TrueNorth for optical flow estimation from event-based sensors, combining low power consumption with high temporal resolution.

Findings

Achieves 11% average endpoint error on test recordings.

Operates at under 80mW power budget.

Uses a variant of the Barlow Levick method for optical flow.

Abstract

This paper describes a fully spike-based neural network for optical flow estimation from Dynamic Vision Sensor data. A low power embedded implementation of the method which combines the Asynchronous Time-based Image Sensor with IBM's TrueNorth Neurosynaptic System is presented. The sensor generates spikes with sub-millisecond resolution in response to scene illumination changes. These spike are processed by a spiking neural network running on TrueNorth with a 1 millisecond resolution to accurately determine the order and time difference of spikes from neighboring pixels, and therefore infer the velocity. The spiking neural network is a variant of the Barlow Levick method for optical flow estimation. The system is evaluated on two recordings for which ground truth motion is available, and achieves an Average Endpoint Error of 11% at an estimated power budget of under 80mW for the sensor and computation.