Deep Learning supported gait analysis
Gait analysis is important for the assessment of clinical endpoints for the therapy of e.g. Morbus Parkinson or Multiple Sclerosis (MS).
However, the current procedure requires a gait analysis lab setup or exhaustive manual annotation work.
Our aim is to develop a tool for marker-free video-based gait analysis under real-world conditions.
Therefore, we strive to employ modern techniques of Computer Vision and Deep Learning,
to extract the clinically relevant gait parameters with a high level of automation.