HDRT: A Large-Scale Dataset for Infrared-Guided HDR Imaging

TL;DR

This paper introduces HDRT, a large-scale dataset of HDR and infrared images, and proposes HDRTNet, a deep learning method that fuses IR and SDR images to produce high-quality HDR images, addressing limitations of traditional HDR techniques.

Contribution

The paper presents the first comprehensive HDR and IR image dataset and a novel neural network for IR-guided HDR imaging, enabling improved image quality and broader research applications.

Findings

HDRTNet outperforms state-of-the-art methods in quality metrics.

The HDRT dataset covers diverse conditions across seasons and cities.

Significant improvements in HDR image reconstruction quality.

Abstract

Capturing images with enough details to solve imaging tasks is a long-standing challenge in imaging, particularly due to the limitations of standard dynamic range (SDR) images which often lose details in underexposed or overexposed regions. Traditional high dynamic range (HDR) methods, like multi-exposure fusion or inverse tone mapping, struggle with ghosting and incomplete data reconstruction. Infrared (IR) imaging offers a unique advantage by being less affected by lighting conditions, providing consistent detail capture regardless of visible light intensity. In this paper, we introduce the HDRT dataset, the first comprehensive dataset that consists of HDR and thermal IR images. The HDRT dataset comprises 50,000 images captured across three seasons over six months in eight cities, providing a diverse range of lighting conditions and environmental contexts. Leveraging this dataset, we propose HDRTNet, a novel deep neural method that fuses IR and SDR content to generate HDR images. Extensive experiments validate HDRTNet against the state-of-the-art, showing substantial quantitative and qualitative quality improvements. The HDRT dataset not only advances IR-guided HDR imaging but also offers significant potential for broader research in HDR imaging, multi-modal fusion, domain transfer, and beyond. The dataset is available at https://huggingface.co/datasets/jingchao-peng/HDRTDataset.