In this project we investigate approaches for unsupervised segmentation of interaction sequences based on multimodal data. The proposed procedure estimates segment borders across all modalities in a single pass.
We develop a 3D VR serious game to integrate valid haptic performance testing in highly adaptable and motivating virtual game scenarios. One goal is to offer stroke patients who need to improve or relearn their proprioceptive and tactile abilities an appealing and motivating tool. It allows them to monitor their progress in a more detailed manner than traditional methods, to select a suitable level of task difficulty, and potentially to improve their self-efficacy. For that purpose we combined multiple devices to a hardware framework creating a touchable virtual world. A set of scenarios is designed where players encounter different haptic discrimination or manipulation challenges in order to evaluate detailed hand and finger movements.
Learning to play piano is a highly demanding activity, which is characterized by a high mental load. To reduce the mental load and, by doing so, to accelerate the learning process, we pursue to use exoskeleton Dexmo.
We are developing a Human-Brain inspired Artificial Intelligence agent capable of producing complex purposeful actions in computer-game simulated environments. Last year, our CITEC team took the first place in the Microsoft Minecraftcompetitionhttps://youtu.be/aqUzh_jHSpY?t=1159.If you would like to know more about this thesis/project opportunity, please contact: Dr. Andrew Melnik <andrew.melnik@uni-bielefeld.de> CITEC-3.308
---------------------------------------------------------------------
read more »
In this project we investigate approaches for unsupervised segmentation of interaction sequences based on multimodal data. The proposed procedure estimates segment borders across all modalities in a single pass.
To download the CAD models of the MHSB please use the following link: 
We develop a 3D VR serious game to integrate valid haptic performance testing in highly adaptable and motivating virtual game scenarios. One goal is to offer stroke patients who need to improve or relearn their proprioceptive and tactile abilities an appealing and motivating tool. It allows them to monitor their progress in a more detailed manner than traditional methods, to select a suitable level of task difficulty, and potentially to improve their self-efficacy. For that purpose we combined multiple devices to a hardware framework creating a touchable virtual world. A set of scenarios is designed where players encounter different haptic discrimination or manipulation challenges in order to evaluate detailed hand and finger movements.
Learning to play piano is a highly demanding activity, which is characterized by a high mental load. To reduce the mental load and, by doing so, to accelerate the learning process, we pursue to use exoskeleton Dexmo.
