Multi-sensor based segmentation of human manipulation tasks

Objectives:

- Dataset acquisition performed to the HANDLE project partner: University of Hamburg - Technical Aspects of Multimodal Systems (TAMS) Lab.

- To acquire multimodal-data about dexterous manipualtion tasks in different scenarios such as: instrumented Rubik-cube, pick-up and "slide"; pick-up and "slide" of long but light piece/block; pick-up a ball-pen with "click-mechanics"; grasp TV-remote-control (or WiiMote or iPhone etc.).

- The goal of the multi-modal dataset acquisition was to understand and extract a basic set of finger and hand movement patterns, which can then be combined to perform a complete manipulation task, and which can be transferred to control robotic hand.

Outputs:

- Datasets of a human demonstrator performing different dexterous manipulation tasks: instrumented Rubik-cube, pick-up and "slide"; pick-up and "slide" of long but light piece/block; pick-up a ball-pen with "click-mechanics"; grasp TV-remote-control (or WiiMote or iPhone etc.). The hand of the human subject was equipped with a Cyberglove, several Polhemus Liberty magnetic tracker sensors and Tekscan tactile sensing grasp array. The demonstrations were also recorded using monocular and stereo RGB cameras. - Request access to HANDLE project database here. Sample dataset video: (to be added)

- Publication by the HANDLE partner University of Hamburg - Technical Aspects of Multimodal Systems (TAMS) Lab.: Hendrich, N.; Klimentjew, D.; Jianwei Zhang, "Multi-sensor based segmentation of human manipulation tasks" Multisensor Fusion and Integration for Intelligent Systems (MFI), 2010 IEEE Conference on , vol., no., pp.223,229, 5-7 Sept. 2010. [PDF]

- Publication by the HANDLE partner University of Hamburg - Technical Aspects of Multimodal Systems (TAMS) Lab. : Norman Hendrich, Denis Klimentjew, and Jianwei Zhang. "Multi-sensor Segmentation of Human Grasping and Manipulation Tasks". 5th International Conference on Cognitive Systems, COGSYS (POSTER). Viena, February, 2012. [PDF]

Scholarly Lite is a free theme, contributed to the Drupal Community by More than Themes.