Walker-Independent Features for Gait Recognition from Motion Capture Data
This addresses the challenge of identifying individuals in applications like video surveillance where labeled data for all people is unavailable, though it appears incremental as it builds on existing discriminant analysis methods.
The paper tackles the problem of gait recognition from motion capture data when new identities appear without labeled data, by introducing walker-independent features learned from raw joint coordinates using a modified Fisher Linear Discriminant Analysis with Maximum Margin Criterion, showing that these features can discriminate people not in the training set and work with fewer learning identities than encountered walkers.
MoCap-based human identification, as a pattern recognition discipline, can be optimized using a machine learning approach. Yet in some applications such as video surveillance new identities can appear on the fly and labeled data for all encountered people may not always be available. This work introduces the concept of learning walker-independent gait features directly from raw joint coordinates by a modification of the Fisher Linear Discriminant Analysis with Maximum Margin Criterion. Our new approach shows not only that these features can discriminate different people than who they are learned on, but also that the number of learning identities can be much smaller than the number of walkers encountered in the real operation.