PID: Physics-Informed Diffusion Model for Infrared Image Generation

TL;DR

This paper introduces a physics-informed diffusion model that translates RGB images into infrared images by embedding physical laws, resulting in more realistic infrared images without increasing training complexity.

Contribution

The proposed PID model uniquely incorporates physical constraints into the diffusion process for RGB to infrared translation, improving realism and adherence to infrared physics.

Findings

PID outperforms existing methods in infrared image quality.

The model maintains physical consistency without extra training parameters.

Experimental results validate the effectiveness of physics-informed constraints.

Abstract

Infrared imaging technology has gained significant attention for its reliable sensing ability in low visibility conditions, prompting many studies to convert the abundant RGB images to infrared images. However, most existing image translation methods treat infrared images as a stylistic variation, neglecting the underlying physical laws, which limits their practical application. To address these issues, we propose a Physics-Informed Diffusion (PID) model for translating RGB images to infrared images that adhere to physical laws. Our method leverages the iterative optimization of the diffusion model and incorporates strong physical constraints based on prior knowledge of infrared laws during training. This approach enhances the similarity between translated infrared images and the real infrared domain without increasing extra training parameters. Experimental results demonstrate that PID significantly outperforms existing state-of-the-art methods. Our code is available at https://github.com/fangyuanmao/PID.