GridFormer: Residual Dense Transformer with Grid Structure for Image Restoration in Adverse Weather Conditions

TL;DR

GridFormer is a transformer-based image restoration framework with a grid structure and residual dense transformer blocks, achieving state-of-the-art results in adverse weather conditions such as deraining, dehazing, and snow removal.

Contribution

Introduces a novel GridFormer framework with enhanced attention and residual dense transformer blocks for improved weather-related image restoration.

Findings

Achieves state-of-the-art results on five weather-related image restoration tasks.

Effective in deraining, dehazing, snow removal, and multi-weather restoration.

Provides open-source code and pre-trained models.

Abstract

Image restoration in adverse weather conditions is a difficult task in computer vision. In this paper, we propose a novel transformer-based framework called GridFormer which serves as a backbone for image restoration under adverse weather conditions. GridFormer is designed in a grid structure using a residual dense transformer block, and it introduces two core designs. First, it uses an enhanced attention mechanism in the transformer layer. The mechanism includes stages of the sampler and compact self-attention to improve efficiency, and a local enhancement stage to strengthen local information. Second, we introduce a residual dense transformer block (RDTB) as the final GridFormer layer. This design further improves the network's ability to learn effective features from both preceding and current local features. The GridFormer framework achieves state-of-the-art results on five diverse image restoration tasks in adverse weather conditions, including image deraining, dehazing, deraining \& dehazing, desnowing, and multi-weather restoration. The source code and pre-trained models are available at https://github.com/TaoWangzj/GridFormer.