Ultra-High-Definition Low-Light Image Enhancement: A Benchmark and Transformer-Based Method

TL;DR

This paper introduces a large-scale UHD low-light image dataset, benchmarks existing algorithms, and proposes LLFormer, a transformer-based method that outperforms current state-of-the-art techniques in low-light image enhancement.

Contribution

The paper provides a new UHD dataset for low-light images, conducts comprehensive benchmarking, and presents LLFormer, a novel transformer-based approach with reduced complexity for low-light image enhancement.

Findings

LLFormer outperforms existing methods on benchmark datasets.

Using the benchmark-trained LLIE methods improves downstream tasks.

The new dataset enables large-scale evaluation of LLIE algorithms.

Abstract

As the quality of optical sensors improves, there is a need for processing large-scale images. In particular, the ability of devices to capture ultra-high definition (UHD) images and video places new demands on the image processing pipeline. In this paper, we consider the task of low-light image enhancement (LLIE) and introduce a large-scale database consisting of images at 4K and 8K resolution. We conduct systematic benchmarking studies and provide a comparison of current LLIE algorithms. As a second contribution, we introduce LLFormer, a transformer-based low-light enhancement method. The core components of LLFormer are the axis-based multi-head self-attention and cross-layer attention fusion block, which significantly reduces the linear complexity. Extensive experiments on the new dataset and existing public datasets show that LLFormer outperforms state-of-the-art methods. We also show that employing existing LLIE methods trained on our benchmark as a pre-processing step significantly improves the performance of downstream tasks, e.g., face detection in low-light conditions. The source code and pre-trained models are available at https://github.com/TaoWangzj/LLFormer.