Extraction of temporal motor activity signals from video recordings of neonatal seizures by feature tracking methods based on rigid motion models

This paper presents a new method for tracking features in video. This method estimates the displacement of a feature between two successive frames by minimizing an error function defined in terms of the feature intensities at these frames. Feature tracking relies on a rigid motion model, which allows for translation, rotation and uniform scaling of the feature tracked from one frame to the next. The proposed feature tracking method is used to extract motor activity signals from video recordings of neonatal seizures.