Learning A Spiking Neural Network for Efficient Image Deraining

TL;DR

This paper introduces ESDNet, an efficient spiking neural network designed for image deraining, which leverages spike signal analysis and a novel training strategy to outperform traditional methods with lower energy consumption.

Contribution

We propose a novel spiking residual block and a gradient proxy training strategy to improve SNN performance in image deraining tasks, addressing information loss and training challenges.

Findings

Achieves comparable deraining performance to ANN-based methods.

Reduces energy consumption by 54%.

Effectively detects rain streaks through spike signal analysis.

Abstract

Recently, spiking neural networks (SNNs) have demonstrated substantial potential in computer vision tasks. In this paper, we present an Efficient Spiking Deraining Network, called ESDNet. Our work is motivated by the observation that rain pixel values will lead to a more pronounced intensity of spike signals in SNNs. However, directly applying deep SNNs to image deraining task still remains a significant challenge. This is attributed to the information loss and training difficulties that arise from discrete binary activation and complex spatio-temporal dynamics. To this end, we develop a spiking residual block to convert the input into spike signals, then adaptively optimize the membrane potential by introducing attention weights to adjust spike responses in a data-driven manner, alleviating information loss caused by discrete binary activation. By this way, our ESDNet can effectively detect and analyze the characteristics of rain streaks by learning their fluctuations. This also enables better guidance for the deraining process and facilitates high-quality image reconstruction. Instead of relying on the ANN-SNN conversion strategy, we introduce a gradient proxy strategy to directly train the model for overcoming the challenge of training. Experimental results show that our approach gains comparable performance against ANN-based methods while reducing energy consumption by 54%. The code source is available at https://github.com/MingTian99/ESDNet.