Co-FRIEND aims to design a framework for understanding human activities in real environments, through an artificial cognitive vision system, identifying objects and events, and extracting sense from scene observation. It will manage uncertainty and change, and will create analysis meaning.
The aim of this project is to advance the state of the art of cognitive systems by developing a methodology for autonomous and continuous learning. The project will concentrate on structural learning, where relations between components and compositional hierarchies play a central role in object categorization. Such learning is particularly relevant for the interpretation of man-made objects, hence the project will use the recognition of buildings in outdoor scenes as its exemplary application domain.
The aim of this project is to develop methods for analysing 3D tomography images of medium density fibreboards (MDF), a material made of wood commonly used in the industry. The tasks include the segmentation of individual fibres, finding the contact surface between the fibres, determining the amount of adhesive resin, calculating the lumen volume and visualising the results.
The objective of this project is to provide the methods and techniques that enable construction of vision systems that can perform task oriented categorization and recognition of objects and events in the context of an embodied agent. The functionality will enable construction of mobile agents that can interpret the action of humans and interact with the environment for tasks such as fetch and delivery of objects in a realistic domestic setting.
A video showing the combination of several low-level segmenters using a probabilistic high-level model provided by the Bayesian Compositional Hierarchy. The best combination of regions from the evidence soup is selected and classified based on a set of valid facade models. Done together with Arne Kreutzmann
Two videos showing the segmentation of wood fibres. The first video shows the fibre segmentation process, starting with noise removal, following up with lumen detection and finally showing a detected fibre. The second video shows the detection of the adhesive resin.
Two videos showing the interpretation process on table-setting scenes. The first video shows the tracking and object detection module which provides evidence. The second video shows a part of the interpretation process using the KONWERK tool and the final interpretation. Done together with Lothar Hotz
Die Bilder für die Übungsaufgaben für die Vorlesung "Bildverarbeitung" von Prof. Neumann finden Sie hier:
Blatt 1, Aufgabe 2: Bayer filter image
Blatt 3, Aufgabe 4: Salt'n'peppa