Deep learning for visual cognitive systems

Abstract

In this talk, I will present the recent work of the Visual Cognitive Systems Laboratory, a research group at the University of Ljubljana. Our team focuses on both fundamental and practical research, exploring the potential of visually enabled intelligent systems within the realms of computer vision, (deep) machine learning, and cognitive robotics.During the presentation, I will showcase our diverse range of projects, which span several areas. We have made significant contributions to visual object tracking and worked on detection, categorization, and segmentation, with applications extending to crucial domains such as visual inspection for quality control, visual surveillance, robot navigation, and remote sensing. We will also explore some unexpected but exciting avenues, these seemingly outlier topics encompass educational robotics, where we explore the integration of DIY robotics and visual information in educational settings, as well as weather forecasting, where we investigate the potential of computer vision and machine learning techniques in improving meteorological predictions.

About the speaker

Luka Čehovin Zajc received his PhD from the Faculty of Computer and Information Science, University of Ljubljana, Slovenia in 2015. He is working at the Visual Cognitive Systems Laboratory, Faculty of Computer and Information Science, University of Ljubljana, Slovenia as an assistant professor and a researcher. He is a founding member of the VOT Initiative. His research interests include computer vision, robotics, human-computer interaction, and spatial computing.