Data Set Used
— Human motion recognition is traditionally approached by either recognizing basic motions from features derived from video input or by interpreting complex motions by applying a high-level hierarchy of motion primitives. The former method is usually limited to rather simple motions while the latter requires human expert knowledge to build up a suitable… (More)
We present an approach for video based human motion capture using a static multi camera setup. The image data of calibrated video cameras is used to generate dense volumentric reconstructions of a person within the capure volume. The 3d reconstructions are then used to fit a 3d cone model into the data utilizing the Iterative Closest Point (ICP) algorithm.… (More)
Many multi camera based approaches on dense volumet-ric 3d reconstruction depend on the correctly chosen often manually defined reconstruction volume and a reliable fore-ground/background segmentation process. We present an algorithm to automatically calculate the 3d polyhedron of the reconstruction volume seen by all cameras of a calibrated multi camera… (More)
The correct segmentation of articulated motion is an important factor to extract and understand the functional structures of complex, articulated objects. Segmenting such body motion without additional appearance information is still a challenging task, because articulated objects as e.g. the human body are mainly based on fine, connected structures. The… (More)
A.: Adaptive foreground/background segmentation using multiview silhouette fusion. A.: " grabcut " : interactive foreground extraction using iterated graph cuts.