Multi-Scale Locality-Constrained Spatiotemporal Coding for Local Feature Based Human Action Recognition
We propose a Multiscale Locality-Constrained Spatiotemporal Coding (MLSC) method to improve the traditional bag of features (BoF) algorithm which ignores the spatiotemporal relationship of local features for human action recognition in video. To model this spatiotemporal relationship, MLSC involves...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Wiley
2013-01-01
|
| Series: | The Scientific World Journal |
| Online Access: | http://dx.doi.org/10.1155/2013/405645 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832556140226936832 |
|---|---|
| author | Bin Wang Yu Liu Wei Wang Wei Xu Maojun Zhang |
| author_facet | Bin Wang Yu Liu Wei Wang Wei Xu Maojun Zhang |
| author_sort | Bin Wang |
| collection | DOAJ |
| description | We propose a Multiscale Locality-Constrained Spatiotemporal Coding (MLSC) method to improve the traditional bag of features (BoF) algorithm which ignores the spatiotemporal relationship of local features for human action recognition in video. To model this spatiotemporal relationship, MLSC involves the spatiotemporal position of local feature into feature coding processing. It projects local features into a sub space-time-volume (sub-STV) and encodes them with a locality-constrained linear coding. A group of sub-STV features obtained from one video with MLSC and max-pooling are used to classify this video. In classification stage, the Locality-Constrained Group Sparse Representation (LGSR) is adopted to utilize the intrinsic group information of these sub-STV features. The experimental results on KTH, Weizmann, and UCF sports datasets show that our method achieves better performance than the competing local spatiotemporal feature-based human action recognition methods. |
| format | Article |
| id | doaj-art-6a87e95a8883495aa81cfedf59ba0658 |
| institution | Kabale University |
| issn | 1537-744X |
| language | English |
| publishDate | 2013-01-01 |
| publisher | Wiley |
| record_format | Article |
| series | The Scientific World Journal |
| spelling | doaj-art-6a87e95a8883495aa81cfedf59ba06582025-02-03T05:46:17ZengWileyThe Scientific World Journal1537-744X2013-01-01201310.1155/2013/405645405645Multi-Scale Locality-Constrained Spatiotemporal Coding for Local Feature Based Human Action RecognitionBin Wang0Yu Liu1Wei Wang2Wei Xu3Maojun Zhang4College of Information System and Manage, National University of Defense Technology, 109 Deya Road, Changsha, Hunan 410073, ChinaCollege of Information System and Manage, National University of Defense Technology, 109 Deya Road, Changsha, Hunan 410073, ChinaCollege of Information System and Manage, National University of Defense Technology, 109 Deya Road, Changsha, Hunan 410073, ChinaCollege of Information System and Manage, National University of Defense Technology, 109 Deya Road, Changsha, Hunan 410073, ChinaCollege of Information System and Manage, National University of Defense Technology, 109 Deya Road, Changsha, Hunan 410073, ChinaWe propose a Multiscale Locality-Constrained Spatiotemporal Coding (MLSC) method to improve the traditional bag of features (BoF) algorithm which ignores the spatiotemporal relationship of local features for human action recognition in video. To model this spatiotemporal relationship, MLSC involves the spatiotemporal position of local feature into feature coding processing. It projects local features into a sub space-time-volume (sub-STV) and encodes them with a locality-constrained linear coding. A group of sub-STV features obtained from one video with MLSC and max-pooling are used to classify this video. In classification stage, the Locality-Constrained Group Sparse Representation (LGSR) is adopted to utilize the intrinsic group information of these sub-STV features. The experimental results on KTH, Weizmann, and UCF sports datasets show that our method achieves better performance than the competing local spatiotemporal feature-based human action recognition methods.http://dx.doi.org/10.1155/2013/405645 |
| spellingShingle | Bin Wang Yu Liu Wei Wang Wei Xu Maojun Zhang Multi-Scale Locality-Constrained Spatiotemporal Coding for Local Feature Based Human Action Recognition The Scientific World Journal |
| title | Multi-Scale Locality-Constrained Spatiotemporal Coding for Local Feature Based Human Action Recognition |
| title_full | Multi-Scale Locality-Constrained Spatiotemporal Coding for Local Feature Based Human Action Recognition |
| title_fullStr | Multi-Scale Locality-Constrained Spatiotemporal Coding for Local Feature Based Human Action Recognition |
| title_full_unstemmed | Multi-Scale Locality-Constrained Spatiotemporal Coding for Local Feature Based Human Action Recognition |
| title_short | Multi-Scale Locality-Constrained Spatiotemporal Coding for Local Feature Based Human Action Recognition |
| title_sort | multi scale locality constrained spatiotemporal coding for local feature based human action recognition |
| url | http://dx.doi.org/10.1155/2013/405645 |
| work_keys_str_mv | AT binwang multiscalelocalityconstrainedspatiotemporalcodingforlocalfeaturebasedhumanactionrecognition AT yuliu multiscalelocalityconstrainedspatiotemporalcodingforlocalfeaturebasedhumanactionrecognition AT weiwang multiscalelocalityconstrainedspatiotemporalcodingforlocalfeaturebasedhumanactionrecognition AT weixu multiscalelocalityconstrainedspatiotemporalcodingforlocalfeaturebasedhumanactionrecognition AT maojunzhang multiscalelocalityconstrainedspatiotemporalcodingforlocalfeaturebasedhumanactionrecognition |