Scalable Video Visual Analytics SPP1335
We focus on visual analytics of video data in the context of the priority program "Scalable Visual Analytics: Interactive Visual Analysis Systems of Complex Information Spaces" (SPP1335).
"Yesterday the German Chancellor officially declared this years CeBIT open. For the next week his hometown will be the host for the worlds biggest IT fair."
Even for a human these two sentences can represent quite a challenge. The word his in the second sentence probably refers the phrase German Chancellor in the first one as well as the phrase the worlds biggest IT fair might refer to the term CeBIT in the first sentence. If you were given access to some additional information e.g. that these two sentence came from a newpaper article in 2004, you might also infer that the German Chancellor refers to the person Gerhard Schröder and his hometown to the city of Hannover as it was his hometown at that point and has been the host city for the IT fair CeBIT ever since. In addition you may find further evidence that the worlds biggest IT fair indeed refers to the term CeBIT.
We currently research a new approach that incorporates world knowledge in this task known as (Co)Reference Resolution with the aim to find a more accurate and reliable way of making huge collections of data accessible to both humans and machines.
Interactive 3D Modelling
A multitude of computer vision based techniques reconstruct and model 3D objects or scenes from photographs or video footage captured in 2D (monocular) or 3D (stereo). Many of these techniques and approaches reconstruct objects or scenes with automatic algorithms from image sequences. But would an architect or an engineer call the resulting 3D reconstruction a CAD-model? Are users able to reconstruct models of real world objects in such a way that they will be able to translate them back to real world replicas? The answer to these questions is mostly – No – but why?
The development of image recognition systems is a complex and highly specialized process, which requires expert knowledge in areas of image recognition and machine learning. Although image recognition systems have a large potential for future applications in everyday life, their application, as of today, is limited by the lack of access to appropriate development tools.