Exploring the Potentials of Spiking Neural Networks for Image Deraining

TL;DR

This paper introduces a novel Spiking Neural Network approach for image deraining, utilizing the Visual LIF neuron to improve spatial understanding and multi-scale representation, achieving superior results with lower energy consumption.

Contribution

The study proposes the Visual LIF neuron and associated modules, enabling effective hierarchical multi-scale learning in SNNs for image deraining, which was previously underexplored.

Findings

Outperforms state-of-the-art SNN deraining methods

Achieves significant energy efficiency, using only 13% of their energy

Demonstrates strong results across five benchmark datasets

Abstract

Biologically plausible and energy-efficient frameworks such as Spiking Neural Networks (SNNs) have not been sufficiently explored in low-level vision tasks. Taking image deraining as an example, this study addresses the representation of the inherent high-pass characteristics of spiking neurons, specifically in image deraining and innovatively proposes the Visual LIF (VLIF) neuron, overcoming the obstacle of lacking spatial contextual understanding present in traditional spiking neurons. To tackle the limitation of frequency-domain saturation inherent in conventional spiking neurons, we leverage the proposed VLIF to introduce the Spiking Decomposition and Enhancement Module and the lightweight Spiking Multi-scale Unit for hierarchical multi-scale representation learning. Extensive experiments across five benchmark deraining datasets demonstrate that our approach significantly outperforms state-of-the-art SNN-based deraining methods, achieving this superior performance with only 13\% of their energy consumption. These findings establish a solid foundation for deploying SNNs in high-performance, energy-efficient low-level vision tasks.