The recent trend in action recognition is towards larger datasets, an increasing number of action classes and larger visual vocabularies. State-of-the-art human action classification in challenging video data is currently based on a bag-of-visual-words pipeline in which space-time features are aggregated globally to form a histogram. The strategies chosen to sample features and construct a visual vocabulary are critical to performance, in fact often dominating performance. In this work we provide a critical evaluation of various approaches to building a vocabulary and show that good practises do have a significant impact. By subsampling and partitioning features strategically, we are able to achieve state-of-the-art results on 5 major action recognition datasets using relatively small visual vocabularies.
@article{arxiv.1405.7545,
title = {Feature sampling and partitioning for visual vocabulary generation on large action classification datasets},
author = {Michael Sapienza and Fabio Cuzzolin and Philip H. S. Torr},
journal= {arXiv preprint arXiv:1405.7545},
year = {2014}
}