Dual-Branch Remote Sensing Infrared Image Super-Resolution

TL;DR

This paper introduces a dual-branch system for infrared image super-resolution that combines local transformer-based and global modeling approaches, achieving improved image quality in challenge evaluations.

Contribution

It presents a novel dual-branch architecture with test-time local conversion and ensemble techniques, enhancing infrared super-resolution performance.

Findings

Fused output outperforms single branches in PSNR, SSIM, and overall score.

Explicit complementarity between local transformer and global modeling improves results.

Method achieves top performance on the NTIRE 2026 Infrared Image Super-Resolution Challenge.

Abstract

Remote sensing infrared image super-resolution aims to recover sharper thermal observations from low-resolution inputs while preserving target contours, scene layout, and radiometric stability. Unlike visible-image super-resolution, thermal imagery is weakly textured and more sensitive to unstable local sharpening, which makes complementary local and global modeling especially important. This paper presents our solution to the NTIRE 2026 Infrared Image Super-Resolution Challenge, a dual-branch system that combines a HAT-L branch and a MambaIRv2-L branch. The inference pipeline applies test-time local conversion on HAT, eight-way self-ensemble on MambaIRv2, and fixed equal-weight image-space fusion. We report both the official challenge score and a reproducible evaluation on 12 synthetic times-four thermal samples derived from Caltech Aerial RGB-Thermal, on which the fused output outperforms either single branch in PSNR, SSIM, and the overall Score. The results suggest that infrared super-resolution benefits from explicit complementarity between locally strong transformer restoration and globally stable state-space modeling.