Learning Transferable Distance Functions for Human Action Recognition

Abstract

Learning-based approaches for human action recognition often rely on large training sets. Most of these approaches do not perform well when only a few training samples are available. In this chapter,we consider the problem of human ac- tion recognition from a single clip per action. Each clip contains at most 25 frames. Using a patch based motion descriptor and matching scheme, we can achieve promising results on three different action datasetswith a single clip as the template. Our results are comparable to previously published results using much larger train- ing sets.We also present a method for learning a transferable distance function for these patches. The transferable distance function learning extracts generic knowl- edge of patch weighting from previous training sets, and can be applied to videos of new actions without further learning. Our experimental results show that the trans- ferable distance function learning not only improves the recognition accuracy of the single clip action recognition, but also significantly enhances the efficiency of the matching scheme.