The state-of-the-art in computer vision has advanced by leaps and bounds in the last ten years. Robust and reliable systems, made possible by technological advances as well as by the availability of massive quantities of visual data, are able to perform tasks undreamed of only a few years ago. Autonomous driving systems, video surveillance, real time text translation, and image captioning are just a few examples of the applications that are being radically transformed by deep learning applied to computer vision.
Deix is in a unique position to offer deep learning solutions to advanced vision problems. Members of the Scientific Committee are recognized leaders in the field of object recognition and segmentation, as well as video surveillance, self-supervised learning, domain adaptation, and the emerging topics of incremental and contrastive learning.
A specialty of Deix is our holistic approach to problems with significant computer vision components to them. With our diverse portfolio of expertise, we are able to analyze new challenges not only from the perspective of the core vision problem, but also from other perspectives such as the (often hidden) optimization or sensor fusion problems hidden in the broader problem context.
In all of this, Deix brings decades of experience with computer vision, deep learning, optimization and sensor fusion to bear on real-world vision problems.