Vision Transformer with Key-select Routing Attention for Single Image Dehazing

TL;DR

This paper introduces Ksformer, a novel vision transformer model that employs key-select routing attention and frequency processing to improve single image dehazing performance.

Contribution

The paper proposes MKRA for selective attention and LFPM for high-frequency enhancement, advancing single image dehazing techniques.

Findings

Outperforms existing dehazing methods in tests.

Effective multi-scale, multi-channel attention mechanism.

Enhanced high-frequency feature processing.

Abstract

We present Ksformer, utilizing Multi-scale Key-select Routing Attention (MKRA) for intelligent selection of key areas through multi-channel, multi-scale windows with a top-k operator, and Lightweight Frequency Processing Module (LFPM) to enhance high-frequency features, outperforming other dehazing methods in tests.