Traditional Computer Vision in the Deep Learning Era

The ability to extract information from images is a cornerstone in the development of intelligent robotic solutions. Over the past 50 years, we have witnessed a shift from traditional geometric computer vision to deep learning-based approaches. Recently, this transition has accelerated rapidly towards AI-based methods, raising the question of whether there is still room in industry and research for non-data-driven approaches. This workshop aims to bring together experts from both traditional computer vision and deep learning fields to discuss this question. Specifically, deep learning and traditional computer vision experts will be invited to present cutting-edge AI-based solutions and applications, as well as to highlight how traditional computer vision can still make a significant impact in specific domains. The goal is to foster a discussion and determine the role, if any, that traditional computer vision still plays in the era of deep learning.