Automated processing of motion capture data

Through this master's project on motion capture, the student will acquire good knowledge of technologies for tracking human movement, as well as advanced skills in analysing and processing motion data.

Optical infrared motion capture (mocap) technology enables tracking of body movement with high precision. It is widely used in the animation industry (movies, gaming), and also in research within sports, medical rehabilitation, and other areas. However, the technology suffers from certain weaknesses: Sometimes, markers are lost from the tracking for short periods, causing gaps in the data. The system may misidentify a single marker, or even swap the identity of two markers in the middle of the mocap recording

The student choosing this master's project will investigate automated post-processing of motion capture data with the goal of minimising manual post-processing of data. The tasks involve improving current methods for gap-filling of missing marker data and other error corrections, and automated marker labelling of full-body motion capture data.

Suggested qualifications:

Matlab / MoCap Toolbox, Motion capture, Machine learning, Biokinematics. 


Emneord: Motion capture, machine learning, animation
Publisert 31. mai 2017 13:49 - Sist endret 31. mai 2017 13:49


Omfang (studiepoeng)