The First Challenge on Remote Sensing Infrared Image Super-Resolution at NTIRE 2026: Benchmark Results and Method Overview

2026-04-23

AI Summary

This work addresses the problem of ×4 super-resolution for low-resolution infrared images generated via bicubic downscaling in remote sensing scenarios by organizing the first NTIRE 2026 Challenge on Remote Sensing Infrared Image Super-Resolution. To establish a standardized benchmark, the study introduces the first dedicated dataset and evaluation protocol tailored to this task, effectively filling a critical gap in the field. The challenge leverages deep learning approaches specifically designed to account for the unique characteristics of infrared imagery, with quantitative performance assessed using metrics such as PSNR. Attracting 115 participants and 13 teams submitting valid solutions, the competition has yielded a comprehensive overview of representative methods and a performance leaderboard, thereby providing a solid foundation and practical reference for future research.

Technology Category

• Computer Vision: Remote Sensing / Geospatial AI
• Intelligent Robots: Multimodal Perception & Sensor Fusion
• Machine Learning: Multi-instance/Multi-view Learning

Application Category

• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics
• Systems and Infrastructure for Web, Mobile and WoT: Web performance, measurement, and characterization
• Responsible Web: Data and user privacy-enhancing technologies for the Web

Abstract

This paper presents the NTIRE 2026 Remote Sensing Infrared Image Super-Resolution (x4) Challenge, one of the associated challenges of NTIRE 2026. The challenge aims to recover high-resolution (HR) infrared images from low-resolution (LR) inputs generated through bicubic downsampling with a x4 scaling factor. The objective is to develop effective models or solutions that achieve state-of-the-art performance for infrared image SR in remote sensing scenarios. To reflect the characteristics of infrared data and practical application needs, the challenge adopts a single-track setting. A total of 115 participants registered for the competition, with 13 teams submitting valid entries. This report summarizes the challenge design, dataset, evaluation protocol, main results, and the representative methods of each team. The challenge serves as a benchmark to advance research in infrared image super-resolution and promote the development of effective solutions for real-world remote sensing applications.

Problem

• infrared image super-resolution
• remote sensing
• low-resolution to high-resolution
• x4 scaling
• bicubic downsampling

Innovation

• infrared image super-resolution
• remote sensing
• NTIRE challenge
• x4 scaling
• benchmark dataset