Action recognition by learning mid-level motion features

This paper presents a method for human action recognition based on patterns of motion. Previous approaches to action recognition use either local features describing small patches or large-scale features describing the entire human figure. We develop a method constructing mid-level motion features which are built from low-level optical flow information. These features are focused on local regions of the image sequence and are created using a variant of AdaBoost. These features are tuned to discriminate between different classes of action, and are efficient to compute at run-time. A battery of classifiers based on these mid-level features is created and used to classify input sequences. State-of-the-art results are presented on a variety of standard datasets.

1. Introduction

In this paper we address the problem of human action recognition from image sequences. We aim to develop algorithms which can recognize low-level actions such as walking, running, or hand clapping from input video sequences. A reliable solution to this problem would allow for a variety of applications such as automated surveillance, human-computer interaction, and video retrieval and search.