Human action recognition via skeletal and depth based feature fusion

Li, Meng, Leung, Howard and Shum, Hubert P. H. (2016) Human action recognition via skeletal and depth based feature fusion. In: Motion in Games 2016, 10th - 12th October 2016, San Francisco.

[img]
Preview
Text (Conference paper)
mig2016action[1].pdf - Published Version
Available under License Creative Commons Attribution 4.0.

Download (543kB) | Preview
Official URL: http://dx.doi.org/10.1145/2994258.2994268

Abstract

This paper addresses the problem of recognizing human actions captured with depth cameras. Human action recognition is a challenging task as the articulated action data is high dimensional in both spatial and temporal domains. An effective approach to handle this complexity is to divide human body into different body parts according to human skeletal joint positions, and performs recognition based on these part-based feature descriptors. Since different types of features could share some similar hidden structures, and different actions may be well characterized by properties common to all features (sharable structure) and those specific to a feature (specific structure), we propose a joint group sparse regression-based learning method to model each action. Our method can mine the sharable and specific structures among its part-based multiple features meanwhile imposing the importance of these part-based feature structures by joint group sparse regularization, in favor of discriminative part-based feature structure
selection. To represent the dynamics and appearance of the human body parts, we employ part-based multiple features extracted from skeleton and depth data respectively. Then, using the group sparse
regularization techniques, we have derived an algorithm for mining the key part-based features in the proposed learning framework.
The resulting features derived from the learnt weight matrices are more discriminative for multi-task classification. Through extensive experiments on three public datasets, we demonstrate that our approach outperforms existing methods.

Item Type: Conference or Workshop Item (Paper)
Uncontrolled Keywords: action recognition, regularization, feature fusion, group sparse
Subjects: G400 Computer Science
Department: Faculties > Engineering and Environment > Computer and Information Sciences
Depositing User: Hubert P. H. Shum
Date Deposited: 01 Nov 2016 14:57
Last Modified: 05 Sep 2017 09:55
URI: http://nrl.northumbria.ac.uk/id/eprint/28250

Actions (login required)

View Item View Item

Downloads

Downloads per month over past year

View more statistics


Policies: NRL Policies | NRL University Deposit Policy | NRL Deposit Licence