Multi-Temporal Depth Motion Maps-Based Local Binary Patterns for 3-D Human Action Recognition
This paper presents a local spatio-temporal descriptor for action recognistion from depth video sequences, which is capable of distinguishing similar actions as well as coping with different speeds of actions. This descriptor is based on three processing stages. In the first stage, the shape and mot...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2017-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/8055546/ |
| _version_ | 1819172055745036288 |
|---|---|
| author | Chen Chen Mengyuan Liu Hong Liu Baochang Zhang Jungong Han Nasser Kehtarnavaz |
| author_facet | Chen Chen Mengyuan Liu Hong Liu Baochang Zhang Jungong Han Nasser Kehtarnavaz |
| author_sort | Chen Chen |
| collection | DOAJ |
| description | This paper presents a local spatio-temporal descriptor for action recognistion from depth video sequences, which is capable of distinguishing similar actions as well as coping with different speeds of actions. This descriptor is based on three processing stages. In the first stage, the shape and motion cues are captured from a weighted depth sequence by temporally overlapped depth segments, leading to three improved depth motion maps (DMMs) compared with the previously introduced DMMs. In the second stage, the improved DMMs are partitioned into dense patches, from which the local binary patterns histogram features are extracted to characterize local rotation invariant texture information. In the final stage, a Fisher kernel is used for generating a compact feature representation, which is then combined with a kernel-based extreme learning machine classifier. The developed solution is applied to five public domain data sets and is extensively evaluated. The results obtained demonstrate the effectiveness of this solution as compared with the existing approaches. |
| first_indexed | 2024-12-22T20:01:06Z |
| format | Article |
| id | doaj.art-34ded9350cbf41ca98ef94028a2751c2 |
| institution | Directory Open Access Journal |
| issn | 2169-3536 |
| language | English |
| last_indexed | 2024-12-22T20:01:06Z |
| publishDate | 2017-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj.art-34ded9350cbf41ca98ef94028a2751c22022-12-21T18:14:17ZengIEEEIEEE Access2169-35362017-01-015225902260410.1109/ACCESS.2017.27590588055546Multi-Temporal Depth Motion Maps-Based Local Binary Patterns for 3-D Human Action RecognitionChen Chen0https://orcid.org/0000-0003-3957-7061Mengyuan Liu1Hong Liu2Baochang Zhang3Jungong Han4Nasser Kehtarnavaz5Center for Research in Computer Vision, University of Central Florida, Orlando, FL, USAShenzhen Graduate School, Peking University, Beijing, ChinaShenzhen Graduate School, Peking University, Beijing, ChinaBeihang University, Beijing, ChinaSchool of Computing and Communications, Lancaster University, Lancaster, U.K.Department of Electrical Engineering, University of Texas at Dallas, Richardson, TX, USAThis paper presents a local spatio-temporal descriptor for action recognistion from depth video sequences, which is capable of distinguishing similar actions as well as coping with different speeds of actions. This descriptor is based on three processing stages. In the first stage, the shape and motion cues are captured from a weighted depth sequence by temporally overlapped depth segments, leading to three improved depth motion maps (DMMs) compared with the previously introduced DMMs. In the second stage, the improved DMMs are partitioned into dense patches, from which the local binary patterns histogram features are extracted to characterize local rotation invariant texture information. In the final stage, a Fisher kernel is used for generating a compact feature representation, which is then combined with a kernel-based extreme learning machine classifier. The developed solution is applied to five public domain data sets and is extensively evaluated. The results obtained demonstrate the effectiveness of this solution as compared with the existing approaches.https://ieeexplore.ieee.org/document/8055546/Action recognitiondepth motion mapsELM classifierlocal binary patternsfisher kernel |
| spellingShingle | Chen Chen Mengyuan Liu Hong Liu Baochang Zhang Jungong Han Nasser Kehtarnavaz Multi-Temporal Depth Motion Maps-Based Local Binary Patterns for 3-D Human Action Recognition IEEE Access Action recognition depth motion maps ELM classifier local binary patterns fisher kernel |
| title | Multi-Temporal Depth Motion Maps-Based Local Binary Patterns for 3-D Human Action Recognition |
| title_full | Multi-Temporal Depth Motion Maps-Based Local Binary Patterns for 3-D Human Action Recognition |
| title_fullStr | Multi-Temporal Depth Motion Maps-Based Local Binary Patterns for 3-D Human Action Recognition |
| title_full_unstemmed | Multi-Temporal Depth Motion Maps-Based Local Binary Patterns for 3-D Human Action Recognition |
| title_short | Multi-Temporal Depth Motion Maps-Based Local Binary Patterns for 3-D Human Action Recognition |
| title_sort | multi temporal depth motion maps based local binary patterns for 3 d human action recognition |
| topic | Action recognition depth motion maps ELM classifier local binary patterns fisher kernel |
| url | https://ieeexplore.ieee.org/document/8055546/ |
| work_keys_str_mv | AT chenchen multitemporaldepthmotionmapsbasedlocalbinarypatternsfor3dhumanactionrecognition AT mengyuanliu multitemporaldepthmotionmapsbasedlocalbinarypatternsfor3dhumanactionrecognition AT hongliu multitemporaldepthmotionmapsbasedlocalbinarypatternsfor3dhumanactionrecognition AT baochangzhang multitemporaldepthmotionmapsbasedlocalbinarypatternsfor3dhumanactionrecognition AT jungonghan multitemporaldepthmotionmapsbasedlocalbinarypatternsfor3dhumanactionrecognition AT nasserkehtarnavaz multitemporaldepthmotionmapsbasedlocalbinarypatternsfor3dhumanactionrecognition |