Bio: Michael Felsberg is a Full Professor at Linköping University's Computer Vision Laboratory since 2008, where he has established himself as a leading researcher in artificial visual systems (AVS). After receiving his PhD from Kiel University, Germany, in 2002, and his docent degree from Linköping University in 2005, he received the DAGM Olympus award in 2005. Felsberg is a fellow of the IAPR, ELLIS, and AAIA, and is a key figure in WASP (Wallenberg AI, Autonomous Systems and Software Program) since its inception. Felsberg has developed expertise spanning three-dimensional computer vision, computational imaging, object detection, tracking and recognition, robot vision, and autonomous systems.
Felsberg's current research addresses some of the most challenging problems in computer vision, including visual object tracking, video object and instance segmentation, and point cloud classification and registration. His work on efficient machine learning techniques for incremental, few-shot, and long-tailed learning settings is particularly relevant for young researchers grappling with real-world AI deployment challenges where data may be limited or continuously evolving. Recent initiatives in his laboratory also tackle problems related to climate change and discovery of new materials, demonstrating how fundamental computer vision research can contribute to pressing global issues.
NAIM 2025 - talk:
Linköping University has recently decided in inaugurate a center of excellence in AI, AI4X, led by Fredrik Heintz, and closely connected to the Computer Vision Laboratory (CVL). During the process of shaping the profile of the center, three different perspectives to AI research have been identified: a) research within AI (AI methods, the main focus of AI4X), b) research with AI (AI for science), and c) research about AI (e.g. regarding societal consequences).
The field of computer vision is a sub area of AI, thus belongs to perspective a), and had major contributions to machine learning (ML) methodology, such as convolutional neural networks. However, many of the ML technologies developed in other domains have successfully been used to solve vision problems, in particular from natural language processing, making it also a case of perspective b).
Also, computer vision has a long tradition to relate technology to biological observations as it has its roots in the modeling of the human visual system (HVS). The highly intuitive nature of the HVS makes it difficult for us to understand the myriad of interdisciplinary problems associated with computer vision, but also the impact of their solutions to humans and society, i.e., perspective c).
The research at CVL has a strong focus on theory and methods, but also on their application to problems that are of great societal importance, such as safe autonomus systems, sustainable forestry and agriculture, monitoring of greenhouse gases, classification and monitoring of animals, as well as the discovery of new materials. The choice of research topics is often driven by findings about AI, e.g., the need for fairness, reliability, correctness, and reduction of resources, i.e., by perspective c).
Felsberg's current research addresses some of the most challenging problems in computer vision, including visual object tracking, video object and instance segmentation, and point cloud classification and registration. His work on efficient machine learning techniques for incremental, few-shot, and long-tailed learning settings is particularly relevant for young researchers grappling with real-world AI deployment challenges where data may be limited or continuously evolving. Recent initiatives in his laboratory also tackle problems related to climate change and discovery of new materials, demonstrating how fundamental computer vision research can contribute to pressing global issues.
NAIM 2025 - talk:
Linköping University has recently decided in inaugurate a center of excellence in AI, AI4X, led by Fredrik Heintz, and closely connected to the Computer Vision Laboratory (CVL). During the process of shaping the profile of the center, three different perspectives to AI research have been identified: a) research within AI (AI methods, the main focus of AI4X), b) research with AI (AI for science), and c) research about AI (e.g. regarding societal consequences).
The field of computer vision is a sub area of AI, thus belongs to perspective a), and had major contributions to machine learning (ML) methodology, such as convolutional neural networks. However, many of the ML technologies developed in other domains have successfully been used to solve vision problems, in particular from natural language processing, making it also a case of perspective b).
Also, computer vision has a long tradition to relate technology to biological observations as it has its roots in the modeling of the human visual system (HVS). The highly intuitive nature of the HVS makes it difficult for us to understand the myriad of interdisciplinary problems associated with computer vision, but also the impact of their solutions to humans and society, i.e., perspective c).
The research at CVL has a strong focus on theory and methods, but also on their application to problems that are of great societal importance, such as safe autonomus systems, sustainable forestry and agriculture, monitoring of greenhouse gases, classification and monitoring of animals, as well as the discovery of new materials. The choice of research topics is often driven by findings about AI, e.g., the need for fairness, reliability, correctness, and reduction of resources, i.e., by perspective c).
