AstraSR: Real-World Thermal Super-Resolution with GPT-6 Astra

2026-10-05

AI Summary

This study addresses the absence of high-resolution references and the synthetic-to-real domain shift in real-world thermal image super-resolution by proposing a novel paradigm that leverages state-of-the-art multimodal large models to generate reference images. Methodologically, GPT-6 Astra is employed for conditional generation to construct high-quality, high-resolution references, while a joint loss function integrating pixel, gradient, and perceptual terms is designed to provide direct supervision. Experimental results demonstrate that the proposed approach significantly outperforms seven existing state-of-the-art algorithms in image clarity and structural coherence, effectively achieving precise restoration of intensity, structure, and fine details.

Abstract

Real-world thermal super-resolution (SR) is constrained by limited sensor resolution and the difficulty of obtaining corresponding high-resolution (HR) observations for direct model supervision. Conventional SR methods typically construct training pairs by treating captured thermal images with real-world degradations as HR references and applying predefined degradation to generate synthetic low-resolution (LR) inputs. Such a construction not only introduces a domain gap between synthetic and captured LR observations but also retains acquisition degradations in the supervision. To address this issue, we propose AstraSR, a real-world thermal SR method guided by GPT-6 Astra, a frontier multimodal generative model endowed with emergent and transformative visual capabilities. Specifically, we construct a dataset of image pairs by using captured LR thermal images to condition GPT-based HR reference. We develop a direct generative supervision strategy that learns from captured thermal inputs paired with GPT-generated HR references. Pixel, gradient, and perceptual losses jointly supervise the transfer of intensity patterns, structural boundaries, and visual details from the generated references. Qualitative comparisons with seven existing state-of-the-art real-world SR methods show continuous object contours, distinct structural boundaries, and smooth intensity transitions in the thermal scenes. These results demonstrate that AstraSR outperforms existing real-world SR methods in both thermal clarity and structural coherence.

Problem

• Thermal Super-Resolution
• Real-World Degradation
• Domain Gap
• High-Resolution Supervision

Innovation

• Thermal Super-Resolution
• Multimodal Generative Model
• Generative Supervision
• Domain Gap
• Perceptual Loss