Depth-based human activity recognition via multi-level fused features and fast broad learning system

Human activity recognition using depth videos remains a challenging problem while in some applications the available training samples is limited. In this article, we propose a new method for human activity recognition by crafting an integrated descriptor called multi-level fused features for depth s...

Full description

Saved in:
Bibliographic Details
Main Authors: Huang Yao, Mengting Yang, Tiantian Chen, Yantao Wei, Yu Zhang
Format: Article
Language:English
Published: Wiley 2020-02-01
Series:International Journal of Distributed Sensor Networks
Online Access:https://doi.org/10.1177/1550147720907830
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849468998839697408
author Huang Yao
Mengting Yang
Tiantian Chen
Yantao Wei
Yu Zhang
author_facet Huang Yao
Mengting Yang
Tiantian Chen
Yantao Wei
Yu Zhang
author_sort Huang Yao
collection DOAJ
description Human activity recognition using depth videos remains a challenging problem while in some applications the available training samples is limited. In this article, we propose a new method for human activity recognition by crafting an integrated descriptor called multi-level fused features for depth sequences and devising a fast broad learning system based on matrix decomposition for classification. First, the surface normals are computed from original depth maps; the histogram of the surface normal orientations is obtained as a low-level feature by accumulating the contributions from normals, then a high-level feature is acquired by sparse coding and pooling on the aggregation of polynormals. After that, the principal component analysis is applied to the conjunction of the two-level features in order to obtain a low-dimensional and discriminative fused feature. At last, fast broad learning system based on matrix decomposition is proposed to accelerate the training process and enhance the classification results. The recognition results on three benchmark data sets show that our method outperforms the state-of-the-art methods in term of accuracy, especially when the number of training samples is small.
format Article
id doaj-art-ff467caec98943e8b63f4272ead7f9f6
institution Kabale University
issn 1550-1477
language English
publishDate 2020-02-01
publisher Wiley
record_format Article
series International Journal of Distributed Sensor Networks
spelling doaj-art-ff467caec98943e8b63f4272ead7f9f62025-08-20T03:25:40ZengWileyInternational Journal of Distributed Sensor Networks1550-14772020-02-011610.1177/1550147720907830Depth-based human activity recognition via multi-level fused features and fast broad learning systemHuang YaoMengting YangTiantian ChenYantao WeiYu ZhangHuman activity recognition using depth videos remains a challenging problem while in some applications the available training samples is limited. In this article, we propose a new method for human activity recognition by crafting an integrated descriptor called multi-level fused features for depth sequences and devising a fast broad learning system based on matrix decomposition for classification. First, the surface normals are computed from original depth maps; the histogram of the surface normal orientations is obtained as a low-level feature by accumulating the contributions from normals, then a high-level feature is acquired by sparse coding and pooling on the aggregation of polynormals. After that, the principal component analysis is applied to the conjunction of the two-level features in order to obtain a low-dimensional and discriminative fused feature. At last, fast broad learning system based on matrix decomposition is proposed to accelerate the training process and enhance the classification results. The recognition results on three benchmark data sets show that our method outperforms the state-of-the-art methods in term of accuracy, especially when the number of training samples is small.https://doi.org/10.1177/1550147720907830
spellingShingle Huang Yao
Mengting Yang
Tiantian Chen
Yantao Wei
Yu Zhang
Depth-based human activity recognition via multi-level fused features and fast broad learning system
International Journal of Distributed Sensor Networks
title Depth-based human activity recognition via multi-level fused features and fast broad learning system
title_full Depth-based human activity recognition via multi-level fused features and fast broad learning system
title_fullStr Depth-based human activity recognition via multi-level fused features and fast broad learning system
title_full_unstemmed Depth-based human activity recognition via multi-level fused features and fast broad learning system
title_short Depth-based human activity recognition via multi-level fused features and fast broad learning system
title_sort depth based human activity recognition via multi level fused features and fast broad learning system
url https://doi.org/10.1177/1550147720907830
work_keys_str_mv AT huangyao depthbasedhumanactivityrecognitionviamultilevelfusedfeaturesandfastbroadlearningsystem
AT mengtingyang depthbasedhumanactivityrecognitionviamultilevelfusedfeaturesandfastbroadlearningsystem
AT tiantianchen depthbasedhumanactivityrecognitionviamultilevelfusedfeaturesandfastbroadlearningsystem
AT yantaowei depthbasedhumanactivityrecognitionviamultilevelfusedfeaturesandfastbroadlearningsystem
AT yuzhang depthbasedhumanactivityrecognitionviamultilevelfusedfeaturesandfastbroadlearningsystem