Deep Convolutional Neural Networks for Human Action Recognition Using Depth Maps and Postures
- Submitting institution
-
Liverpool John Moores University
- Unit of assessment
- 11 - Computer Science and Informatics
- Output identifier
- 1409
- Type
- D - Journal article
- DOI
-
10.1109/TSMC.2018.2850149
- Title of journal
- IEEE Transactions on Systems Man and Cybernetics: Systems
- Article number
- -
- First page
- 1806
- Volume
- 49
- Issue
- 9
- ISSN
- 2168-2216
- Open access status
- Exception within 3 months of publication
- Month of publication
- July
- Year of publication
- 2018
- URL
-
-
- Supplementary information
-
-
- Request cross-referral to
- -
- Output has been delayed by COVID-19
- No
- COVID-19 affected output statement
- -
- Forensic science
- No
- Criminology
- No
- Interdisciplinary
- No
- Number of additional authors
-
5
- Research group(s)
-
-
- Citation count
- 17
- Proposed double-weighted
- No
- Reserve for an output with double weighting
- No
- Additional information
- This work resulted from an international collaboration among researchers at four universities in the UK, China and Australia, supported jointly by eight research grants, e.g. 61572316 and 61671290 from the National Natural Science Foundation of China. It presents a novel deep learning based human action recognition approach by fusing depth maps and posture data using convolutional neural networks (CNNs), to effectively and accurately classify human actions. The proposed approach was evaluated on three public datasets, e.g. Microsoft action 3-D dataset (MSRAction3D). The results indicate that the approach outperforms state-of-the-art human recognition methods with increased accuracy up to 6.84%.
- Author contribution statement
- -
- Non-English
- No
- English abstract
- -
