Human Gait Parameter Estimation from Video-Based Motion Capture Data

OData support
Kertész Zsolt László
Department of Control Engineering and Information Technology

The development of biomechanics has given the possibility to investigate how different powers affect certain body parts along with the loads caused by these powers and movements. The methods of investigating movements has significantly developed during the centuries starting from the ancient times till today. The main aim of the investigations is to get closer information about the movement of the body parts and the measuring of these figures firstly in case of common movements like walking, running and swimming. Secondly, it examines these motions in case of specific occurence of implants and their displacement along with the moving of the forms of the so called motion captures.

One of the most important domains of these researches are to make the preventive, remedial and rehabitative methods standing close to locomotor system more objective. Due to the exact measurements we can find numerical results so analysing these figures eliminates subjective points.

The main subject of my thesis is to present different methods of examining movements especially proceedings based on optical methods along with the analysing of the resulting data with the help of matlab procedures. These data are loaded with noise. To decrease the level of the noise I have used the so called methods of sliding window and natural cubic spline. The latter one gives possibility along with the smoothing to define the functional results between the measured points.

Eventually, from the data cleaned and completed with the help of the matlab program we can extract the main features of movement. The 3D display makes it possible to visualise all the information we have got during the analyses.

In my thesis along with the theoretical background I would like also to present the completed programs that makes us possible to define the cleaning of the noise, the interpolarity and the measuring of the features of the motions.


Please sign in to download the files of this thesis.