A Spiking Neural Network for Image Segmentation

TL;DR

This paper demonstrates that converting a U-Net architecture to a spiking neural network on the Loihi chip achieves similar segmentation accuracy as traditional neural networks but with over twice the energy efficiency, suitable for scalable neuromorphic vision tasks.

Contribution

It introduces a novel method for converting ANN to SNN with regularized firing rates, optimizing energy efficiency without accuracy loss, and demonstrates practical deployment on neuromorphic hardware.

Findings

SNN achieves similar segmentation accuracy as ANN.

Neuromorphic implementation is over 2x more energy-efficient.

Power savings are achieved with minimal accuracy loss.

Abstract

We seek to investigate the scalability of neuromorphic computing for computer vision, with the objective of replicating non-neuromorphic performance on computer vision tasks while reducing power consumption. We convert the deep Artificial Neural Network (ANN) architecture U-Net to a Spiking Neural Network (SNN) architecture using the Nengo framework. Both rate-based and spike-based models are trained and optimized for benchmarking performance and power, using a modified version of the ISBI 2D EM Segmentation dataset consisting of microscope images of cells. We propose a partitioning method to optimize inter-chip communication to improve speed and energy efficiency when deploying multi-chip networks on the Loihi neuromorphic chip. We explore the advantages of regularizing firing rates of Loihi neurons for converting ANN to SNN with minimum accuracy loss and optimized energy consumption. We propose a percentile based regularization loss function to limit the spiking rate of the neuron between a desired range. The SNN is converted directly from the corresponding ANN, and demonstrates similar semantic segmentation as the ANN using the same number of neurons and weights. However, the neuromorphic implementation on the Intel Loihi neuromorphic chip is over 2x more energy-efficient than conventional hardware (CPU, GPU) when running online (one image at a time). These power improvements are achieved without sacrificing the task performance accuracy of the network, and when all weights (Loihi, CPU, and GPU networks) are quantized to 8 bits.