Thermal Image Super-Resolution Challenge Results - PBVS 2024

This paper outlines the advancements and results of the Fifth Thermal Image Super-Resolution challenge, hosted at the Perception Beyond the Visible Spectrum CVPR 2024 workshop. The challenge employed a novel benchmark cross-spectral dataset consisting of 1000 thermal images, each paired with its corresponding registered RGB image. The challenge featured two tracks: Track-1 focused on Single Thermal Image Super-Resolution with an ×8 upscale factor, while Track-2 extended its evaluation to include both ×8 and ×16 scaling factors, utilizing high-resolution RGB images to guide the super-resolution process for low-resolution thermal images. The participation of over 175 teams highlights the research community’s strong engagement and dedication to enhancing image resolution techniques across both single and cross-spectral methodologies. This year’s challenge sets new benchmarks and provides valuable insights into future directions for research in thermal image super-resolution.

1. Introduction

The field of image super-resolution (SR), particularly focusing on enhancing the resolution of thermal images, has seen notable advancements in recent years. The primary approach involves using deep learning techniques to convert low-resolution (LR) images into high-resolution (HR) counterparts. These methods typically involve training on downsampled HR images that have been artificially augmented with noise and blur to improve the network’s ability to enhance image quality. Despite the prevalence of such methods for visible spectrum images, there is a growing need for specialized SR techniques tailored for the thermal spectrum due to its wide range of applications.