betaCube - Enhancing Training for Climbing by a Self-Calibrating Camera Projection Unit

Frederik Wiehr, Felix Kosmalla, Florian Daiber, Antonio Krüger

In: CHI'16 Extended Abstracts on Human Factors in Computing Systems. ACM International Conference on Human Factors in Computing Systems (CHI-16). ACM International Conference on Human Factors in Computing Systems (CHI-16) May 7-12 San Jose CA United States ACM 2016.


In rock climbing, discussing climbing techniques with others to master a specific route getting practical advice from more experienced climbers is an inherent part of the culture and tradition of the sport. Spatial information, such as the position of holds, as well as learning complex body postures plays a major role in this process. A typical problem that occurs during advising, is an alignment effect when trying to picture orientation-specific knowledge, e.g. explaining how to perform a certain self-climbed move to others. We propose betaCube, a self-calibrating camera-projection unit that features 3D tracking and distortion-free projection. The system enables a life-sized video replay and climbing route creation using augmented reality. We contribute an interface for automatic setup of mobile and distortion free projection, blob detection for climbing holds, as well as an automatic method for extracting planar trackables from artificial climbing walls.

betaCube.pdf (pdf, 7 MB )

German Research Center for Artificial Intelligence
Deutsches Forschungszentrum für Künstliche Intelligenz