Deep Multi-Threshold Spiking-UNet for Image Processing

TL;DR

This paper introduces a novel Spiking-UNet architecture that combines SNNs with U-Net for image processing, achieving high performance and significantly reduced inference time through innovative training and normalization techniques.

Contribution

It presents a new Spiking-UNet model with multi-threshold neurons and a conversion-based training pipeline, enhancing efficiency and accuracy in neuromorphic image processing.

Findings

Achieves comparable performance to non-spiking U-Net in segmentation and denoising.

Reduces inference time by approximately 90% compared to non-fine-tuned models.

Surpasses existing SNN methods in image processing tasks.

Abstract

U-Net, known for its simple yet efficient architecture, is widely utilized for image processing tasks and is particularly suitable for deployment on neuromorphic chips. This paper introduces the novel concept of Spiking-UNet for image processing, which combines the power of Spiking Neural Networks (SNNs) with the U-Net architecture. To achieve an efficient Spiking-UNet, we face two primary challenges: ensuring high-fidelity information propagation through the network via spikes and formulating an effective training strategy. To address the issue of information loss, we introduce multi-threshold spiking neurons, which improve the efficiency of information transmission within the Spiking-UNet. For the training strategy, we adopt a conversion and fine-tuning pipeline that leverage pre-trained U-Net models. During the conversion process, significant variability in data distribution across different parts is observed when utilizing skip connections. Therefore, we propose a connection-wise normalization method to prevent inaccurate firing rates. Furthermore, we adopt a flow-based training method to fine-tune the converted models, reducing time steps while preserving performance. Experimental results show that, on image segmentation and denoising, our Spiking-UNet achieves comparable performance to its non-spiking counterpart, surpassing existing SNN methods. Compared with the converted Spiking-UNet without fine-tuning, our Spiking-UNet reduces inference time by approximately 90\%. This research broadens the application scope of SNNs in image processing and is expected to inspire further exploration in the field of neuromorphic engineering. The code for our Spiking-UNet implementation is available at https://github.com/SNNresearch/Spiking-UNet.