Search papers, labs, and topics across Lattice.
This paper introduces Text2Thermal, a novel framework that synthesizes thermal images from textual descriptions, addressing the challenge of limited thermal datasets and the ambiguity inherent in translating RGB images to thermal outputs. By leveraging thermally grounded captions that encode critical environmental factors, the method explicitly incorporates unobservable radiometric factors, allowing for the generation of accurate thermal imagery without the need for corresponding RGB images. Experiments demonstrate that Text2Thermal achieves state-of-the-art performance in thermal image synthesis, outperforming existing methods while providing enhanced control through text prompts.
Text2Thermal synthesizes thermal images directly from text, achieving state-of-the-art results while eliminating the need for RGB image registration.
Thermal infrared imaging offers reliable perception in darkness and adverse weather, but thermal datasets remain scarce, motivating extensive work on translating abundant RGB images into thermal. Such translation is fundamentally ill-posed as thermal appearance is governed by surface emissivity and object temperature, neither of which is observable in the visible spectrum, so a single RGB image is consistent with many valid thermal outputs. We argue that language offers a natural means of resolving this ambiguity, and propose Text2Thermal, a framework for physics-aware thermal image synthesis from textual priors. Rather than inferring the unobservable radiometric factors from RGB, we supply them explicitly through thermally grounded captions encoding material, weather, time-of-day, and heat-emission state, and adapt a pretrained Stable Diffusion backbone to the thermal domain. Because the radiometric content is determined entirely by the prompt, Text2Thermal synthesizes thermal imagery without requiring a registered RGB image at inference; where spatial guidance is desired, an optional control signal imparts scene geometry without disturbing the prompt-specified radiometry. Experiments on M3FD, FLIR, and FMB show that Text2Thermal achieves state-of-the-art FID among thermal image synthesis methods while offering text-level control that translation-based approaches cannot provide.
