Toward Real-world Infrared Image Super-Resolution: A Unified Autoregressive Framework and Benchmark Dataset

TL;DR

This paper introduces Real-IISR, a novel autoregressive framework for real-world infrared image super-resolution that effectively handles coupled degradations and thermal-structural details, supported by a new benchmark dataset.

Contribution

The paper presents a unified autoregressive model for real-world infrared super-resolution, incorporating thermal priors, a condition-adaptive codebook, and a thermal order loss, along with a new dataset for evaluation.

Findings

Real-IISR outperforms existing methods on the FLIR-IISR dataset.

The framework effectively reconstructs thermal structures and backgrounds.

The dataset enables comprehensive benchmarking of real-world infrared super-resolution.

Abstract

Infrared image super-resolution (IISR) under real-world conditions is a practically significant yet rarely addressed task. Pioneering works are often trained and evaluated on simulated datasets or neglect the intrinsic differences between infrared and visible imaging. In practice, however, real infrared images are affected by coupled optical and sensing degradations that jointly deteriorate both structural sharpness and thermal fidelity. To address these challenges, we propose Real-IISR, a unified autoregressive framework for real-world IISR that progressively reconstructs fine-grained thermal structures and clear backgrounds in a scale-by-scale manner via thermal-structural guided visual autoregression. Specifically, a Thermal-Structural Guidance module encodes thermal priors to mitigate the mismatch between thermal radiation and structural edges. Since non-uniform degradations typically induce quantization bias, Real-IISR adopts a Condition-Adaptive Codebook that dynamically modulates discrete representations based on degradation-aware thermal priors. Also, a Thermal Order Consistency Loss enforces a monotonic relation between temperature and pixel intensity, ensuring relative brightness order rather than absolute values to maintain physical consistency under spatial misalignment and thermal drift. We build FLIR-IISR, a real-world IISR dataset with paired LR-HR infrared images acquired via automated focus variation and motion-induced blur. Extensive experiments demonstrate the promising performance of Real-IISR, providing a unified foundation for real-world IISR and benchmarking. The dataset and code are available at: https://github.com/JZD151/Real-IISR.