New Thermo-VL model integrates thermal imaging with language models

Researchers have developed Thermo-VL, a new vision-language model designed to process thermal infrared imagery alongside standard RGB data. This model aims to improve performance in low-light conditions by leveraging the complementary scene structure captured by thermal sensors. Thermo-VL integrates a trainable thermal encoder with a frozen language model backbone, using a novel fusion module to condition thermal features on both text and RGB context. AI

IMPACT Enhances vision-language model capabilities for low-light and cross-spectrum reasoning tasks.

RANK_REASON The cluster contains an academic paper detailing a new model and dataset. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.CV →

• Molmo-7B
• Thermo-VL

• paper
• model release

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →