Action detection in office scene based on deep convolutional neural networks

Yan, Shi-Yang, An, Yu-Di, Smith, Jeremy S and Zhang, Bai-Ling (2016) Action detection in office scene based on deep convolutional neural networks. In: 2016 International Conference on Machine Learning and Cybernetics (ICMLC), 2016-7-10 - 2016-7-13, Jeju, Korea.

Text ICMLC_finalVersion.pdf - Author Accepted Manuscript Download (1MB)

Abstract

In many scenarios, a persons behavior in office environment needs to be monitored and some predefined abnormal actions or activities should be detected and recognized. In this paper, we attempted towards the solution starting from a persons pose with poselets as the basic building blocks. The existed powerful pose representation, i.e., poselets, together with deep convolutional neural networks, are exploited to implement an efficient action recognition system from still images. The system extends poselets detector to region proposal, cascaded with R-CNN for final action detection. Unlike many published work which only emphases on action classification, our system implements multi-task learning with classification and localization of person and the corresponding actions simultaneously, To facilitate our studies, a specially designed action dataset was created. Preliminary experiments demonstrate promising results.

Item Type:	Conference or Workshop Item (Unspecified)
Uncontrolled Keywords:	4603 Computer Vision and Multimedia Computation, 46 Information and Computing Sciences, 4611 Machine Learning, Networking and Information Technology R&D (NITRD), Machine Learning and Artificial Intelligence
Depositing User:	Symplectic Admin
Date Deposited:	25 Aug 2017 14:57
Last Modified:	09 Sep 2024 06:50
DOI:	10.1109/icmlc.2016.7860906
Related URLs:	Publisher
URI:	https://livrepository.liverpool.ac.uk/id/eprint/3008488