Classifying movements using efficient kinematic codes

Efficient codes have been shown to perform well in image and audio classification tasks, but the impact of sparsity—and indeed the entire notion of efficient coding—has not yet been well explored in the context of human movements. This paper tests several coding approaches on a movement classification task and finds that efficient codes for kinematic (joint angle) data perform well for classifying many different types of movements. In particular, the best classification method relied on a sparse coding algorithm combined with a codebook that was tuned to kinematic movement data. The other approaches tested here—sparse coding with a random codebook, and "dense" coding using PCA—provide interesting baseline results and allow us to investigate why sparse codes appear to work well.