Human Action Recognition Using Multilevel Depth Motion Maps
The advent of depth sensors opens up new opportunities for human action recognition by providing depth information. The main purpose of this paper is to present an effective method for human action recognition from depth images. A multilevel frame select sampling (MFSS) method are proposed to genera...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2019-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/8675733/ |
| _version_ | 1819170214176096256 |
|---|---|
| author | Xu Weiyao Wu Muqing Zhao Min Liu Yifeng Lv Bo Xia Ting |
| author_facet | Xu Weiyao Wu Muqing Zhao Min Liu Yifeng Lv Bo Xia Ting |
| author_sort | Xu Weiyao |
| collection | DOAJ |
| description | The advent of depth sensors opens up new opportunities for human action recognition by providing depth information. The main purpose of this paper is to present an effective method for human action recognition from depth images. A multilevel frame select sampling (MFSS) method are proposed to generate three levels of temporal samples from the input depth sequences first. Then, the proposed motion and static mapping (MSM) method is used to obtain the representation of MFSS sequences. After that, this paper exploits the block-based LBP feature extraction approach to extract features information from the MSM. Finally, the fisher kernel representation is applied to aggregate the block features, which is then combined with the kernel-based extreme learning machine classifier. The developed framework is evaluated on three public datasets captured by depth cameras. The experimental results demonstrate the great performance compared with the existing approaches. |
| first_indexed | 2024-12-22T19:31:50Z |
| format | Article |
| id | doaj.art-cd380019d1c648dcb6816d05e8283fc9 |
| institution | Directory Open Access Journal |
| issn | 2169-3536 |
| language | English |
| last_indexed | 2024-12-22T19:31:50Z |
| publishDate | 2019-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj.art-cd380019d1c648dcb6816d05e8283fc92022-12-21T18:15:05ZengIEEEIEEE Access2169-35362019-01-017418114182210.1109/ACCESS.2019.29077208675733Human Action Recognition Using Multilevel Depth Motion MapsXu Weiyao0Wu Muqing1Zhao Min2Liu Yifeng3Lv Bo4Xia Ting5Beijing Laboratory of Advanced Information Networks, Beijing University of Posts and Telecommunications, Beijing, ChinaBeijing Laboratory of Advanced Information Networks, Beijing University of Posts and Telecommunications, Beijing, ChinaBeijing Laboratory of Advanced Information Networks, Beijing University of Posts and Telecommunications, Beijing, ChinaChina Academy of Electronics and Information Technology, Beijing, ChinaChina Academy of Electronics and Information Technology, Beijing, ChinaCollege of Opto-electronic Engineering, Zaozhuang University, Zaozhuang, ChinaThe advent of depth sensors opens up new opportunities for human action recognition by providing depth information. The main purpose of this paper is to present an effective method for human action recognition from depth images. A multilevel frame select sampling (MFSS) method are proposed to generate three levels of temporal samples from the input depth sequences first. Then, the proposed motion and static mapping (MSM) method is used to obtain the representation of MFSS sequences. After that, this paper exploits the block-based LBP feature extraction approach to extract features information from the MSM. Finally, the fisher kernel representation is applied to aggregate the block features, which is then combined with the kernel-based extreme learning machine classifier. The developed framework is evaluated on three public datasets captured by depth cameras. The experimental results demonstrate the great performance compared with the existing approaches.https://ieeexplore.ieee.org/document/8675733/Human action recognitiondepth imageELM classifierfisher kernel |
| spellingShingle | Xu Weiyao Wu Muqing Zhao Min Liu Yifeng Lv Bo Xia Ting Human Action Recognition Using Multilevel Depth Motion Maps IEEE Access Human action recognition depth image ELM classifier fisher kernel |
| title | Human Action Recognition Using Multilevel Depth Motion Maps |
| title_full | Human Action Recognition Using Multilevel Depth Motion Maps |
| title_fullStr | Human Action Recognition Using Multilevel Depth Motion Maps |
| title_full_unstemmed | Human Action Recognition Using Multilevel Depth Motion Maps |
| title_short | Human Action Recognition Using Multilevel Depth Motion Maps |
| title_sort | human action recognition using multilevel depth motion maps |
| topic | Human action recognition depth image ELM classifier fisher kernel |
| url | https://ieeexplore.ieee.org/document/8675733/ |
| work_keys_str_mv | AT xuweiyao humanactionrecognitionusingmultileveldepthmotionmaps AT wumuqing humanactionrecognitionusingmultileveldepthmotionmaps AT zhaomin humanactionrecognitionusingmultileveldepthmotionmaps AT liuyifeng humanactionrecognitionusingmultileveldepthmotionmaps AT lvbo humanactionrecognitionusingmultileveldepthmotionmaps AT xiating humanactionrecognitionusingmultileveldepthmotionmaps |