Robust classification of human actions from 3D data

Abstract

We address the problem of classifying human actions using a single depth sensor camera. In this work, we propose an angular representation to model the relationship between the joints in human skeleton. This representation helps cope with noisy data while enhances both computational efficiency and flexibility. Also, we propose to use Hidden Markov Model (HMM) to recognize temporal motion patterns. The full skeleton formulated in a 60D feature vector is tuned to a 37D feature vector of the most active joints. These features are then fed to the HMM for recognition. We evaluate our classifier on a dataset of 19 classes and 5 indoor scenarios with hundreds of action instances recorded using the Microsoft XBOX Kinect J sensor and achieve an average precision/recall of 91.14%/96.89%.

Recognizing Involuntary Actions from 3D Skeleton Data Using Body States

“…Haynh et al [7] proposed a new method more robust to human scale and changes of position. It was performed by categorizing joints into three classes of stable, active and highly active joints and using angles of 10 important joints and vectors connecting moving joints to stable joints.…”

“…Dynamics and appearance of parts were represented by heterogeneous set of depth and skeleton-based features. Huynh et al [17] proposed a new method more robust to human scale and changes of position. They categorized joints into three classes of stable, active and highly active joints, and utilized angles of 10 important joints and vectors connecting moving joints to stable joints.…”