On the Process of Designing an Activity Recognition System Using Symbolic and Subsymbolic Techniques

摘要

In this paper, we address the problem of human activity classificationfrom videos, giving a special emphasis to feature extraction and good feature selection.Due to the cut down in cameras cost that have been in the last years, these kind ofsystems are becoming popular for their wide application area. Taking a video blobtracker output, a feature extraction process is defined to extract an extensive featureset, that is filtered in a later step to select the best features present. Three differenttype of classifiers are trained with the result feature set and results are shown.