Scaling Up Single Image Dehazing Algorithm by Cross-Data Vision Alignment for Richer Representation Learning and Beyond

TL;DR

This paper introduces a cross-data vision alignment method for single image dehazing that leverages domain adaptation and internal augmentation to improve generalization and performance across diverse datasets.

Contribution

The paper proposes a novel cross-data vision alignment technique with self-supervised internal augmentation to address domain gaps in large-scale dehazing training.

Findings

Significantly outperforms existing dehazing methods.

Effectively reduces domain gap across datasets.

Produces dehazed images closest to real haze-free images.

Abstract

In recent years, deep neural networks tasks have increasingly relied on high-quality image inputs. With the development of high-resolution representation learning, the task of image dehazing has received significant attention. Previously, many methods collect diverse image data for large-scale training to boost the performance on a target scene. Ignoring the domain gap between different data, former de-hazing methods simply adopt multiple datasets for explicit large-scale training, which often makes the methods themselves be violated. To address this problem, we propose a novel method of cross-data vision alignment for richer representation learning to improve the existing dehazing methodology. Specifically, we call for the internal- and external knowledge should be further adapted with a self-supervised manner to fill up the domain gap. By using cross-data external alignment, the datasets inherit samples from different domains that are firmly aligned, making the model learn more robust and generalizable features. By using the internal augmentation method, the model can fully exploit local information within the images, and then obtaining more image details. To demonstrate the effectiveness of our proposed method, we conduct training on the Natural Image Dataset (NID). Experimental results show that our method clearly resolves the domain gap in different dehazing datasets and presents a new pipeline for large-scale training in the dehazing task. Our approach significantly outperforms other advanced methods in dehazing and produces dehazed images that are closest to real haze-free images.