MISFIT-V: Misaligned Image Synthesis and Fusion using Information from Thermal and Visual

TL;DR

MISFIT-V introduces an unsupervised deep learning method combining GANs and cross-attention to improve thermal and visual image fusion for better human detection in challenging environments.

Contribution

It presents a novel unsupervised fusion approach using GANs and cross-attention to handle misalignment and environmental challenges in thermal-visual image fusion.

Findings

Enhanced robustness against misalignment

Improved performance in poor lighting and thermal conditions

Outperforms existing fusion methods

Abstract

Detecting humans from airborne visual and thermal imagery is a fundamental challenge for Wilderness Search-and-Rescue (WiSAR) teams, who must perform this function accurately in the face of immense pressure. The ability to fuse these two sensor modalities can potentially reduce the cognitive load on human operators and/or improve the effectiveness of computer vision object detection models. However, the fusion task is particularly challenging in the context of WiSAR due to hardware limitations and extreme environmental factors. This work presents Misaligned Image Synthesis and Fusion using Information from Thermal and Visual (MISFIT-V), a novel two-pronged unsupervised deep learning approach that utilizes a Generative Adversarial Network (GAN) and a cross-attention mechanism to capture the most relevant features from each modality. Experimental results show MISFIT-V offers enhanced robustness against misalignment and poor lighting/thermal environmental conditions compared to existing visual-thermal image fusion methods.