Multimodal image fusion techniques to create enhanced, information-rich images for computer vision applications

Multimodal image fusion refers to extracting relevant information from two or more images from different modalities and effectively combining them by a fusion rule to reconstruct an enhanced image with richer and complementary information. 

We develop deep learning architectures that allow the fusion of multimodal images to enhance image processing and computer vision applications such as edge detection, remote sensing, semantic segmentation, or object detection.