S3GRN: Structural Similar Stepwise Generative Recognizable Network for Human Action Recognition With Limited Training Data
Human action recognition is a hot topic and it has been applied to various fields. Deep learning is one of the techniques in human action recognition which has achieved good results. However, the task is still challenging due to the less collected samples. In order to address this challenge and impr...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2020-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/9272277/ |
| _version_ | 1819132800731709440 |
|---|---|
| author | Chaobo Li Xulin Shen Hongjun Li |
| author_facet | Chaobo Li Xulin Shen Hongjun Li |
| author_sort | Chaobo Li |
| collection | DOAJ |
| description | Human action recognition is a hot topic and it has been applied to various fields. Deep learning is one of the techniques in human action recognition which has achieved good results. However, the task is still challenging due to the less collected samples. In order to address this challenge and improve the recognition accuracy, the stepwise generative recognizable network is proposed based on the generative adversarial network, which can be used to expand limited training samples and then recognize. Firstly, the stepwise generative recognizable network is designed to combine the function of images generation and recognition for human action. Secondly, the structural similar constraint is introduced to stepwise generative recognizable network, called structural similar stepwise generative recognizable network, which can compare the similarity of generated images with real data to improve quality and diversity of generated images. Finally, the performance of proposed networks is verified by common databases and the self-build database which is collected in daily life. We achieved 97.14%, 94.88% and 99.69% recognition accuracy on MNIST, Weizmann and self-build dataset, respectively. The experimental results show that the combination of generation and recognition can improve the recognition accuracy without abundant training data, and the structural similar constraint not only can improve the quality and diversity of generated images but also perform better in convergence. The structural similar stepwise generative recognizable network reduces the workload of manual collection and solves the problem of lower recognition accuracy for limited training samples, which achieves the characteristics of natural expanded samples. |
| first_indexed | 2024-12-22T09:37:10Z |
| format | Article |
| id | doaj.art-eb0b51291e7744fe9dd474a072a6b08a |
| institution | Directory Open Access Journal |
| issn | 2169-3536 |
| language | English |
| last_indexed | 2024-12-22T09:37:10Z |
| publishDate | 2020-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj.art-eb0b51291e7744fe9dd474a072a6b08a2022-12-21T18:30:48ZengIEEEIEEE Access2169-35362020-01-01821621921623010.1109/ACCESS.2020.30407589272277S3GRN: Structural Similar Stepwise Generative Recognizable Network for Human Action Recognition With Limited Training DataChaobo Li0Xulin Shen1https://orcid.org/0000-0002-4571-175XHongjun Li2https://orcid.org/0000-0001-7500-4979School of Information Science and Technology, Nantong University, Nantong, ChinaSchool of Information Science and Technology, Nantong University, Nantong, ChinaSchool of Information Science and Technology, Nantong University, Nantong, ChinaHuman action recognition is a hot topic and it has been applied to various fields. Deep learning is one of the techniques in human action recognition which has achieved good results. However, the task is still challenging due to the less collected samples. In order to address this challenge and improve the recognition accuracy, the stepwise generative recognizable network is proposed based on the generative adversarial network, which can be used to expand limited training samples and then recognize. Firstly, the stepwise generative recognizable network is designed to combine the function of images generation and recognition for human action. Secondly, the structural similar constraint is introduced to stepwise generative recognizable network, called structural similar stepwise generative recognizable network, which can compare the similarity of generated images with real data to improve quality and diversity of generated images. Finally, the performance of proposed networks is verified by common databases and the self-build database which is collected in daily life. We achieved 97.14%, 94.88% and 99.69% recognition accuracy on MNIST, Weizmann and self-build dataset, respectively. The experimental results show that the combination of generation and recognition can improve the recognition accuracy without abundant training data, and the structural similar constraint not only can improve the quality and diversity of generated images but also perform better in convergence. The structural similar stepwise generative recognizable network reduces the workload of manual collection and solves the problem of lower recognition accuracy for limited training samples, which achieves the characteristics of natural expanded samples.https://ieeexplore.ieee.org/document/9272277/Generative networkrecognizable networkcombinationstructural similar constraint |
| spellingShingle | Chaobo Li Xulin Shen Hongjun Li S3GRN: Structural Similar Stepwise Generative Recognizable Network for Human Action Recognition With Limited Training Data IEEE Access Generative network recognizable network combination structural similar constraint |
| title | S3GRN: Structural Similar Stepwise Generative Recognizable Network for Human Action Recognition With Limited Training Data |
| title_full | S3GRN: Structural Similar Stepwise Generative Recognizable Network for Human Action Recognition With Limited Training Data |
| title_fullStr | S3GRN: Structural Similar Stepwise Generative Recognizable Network for Human Action Recognition With Limited Training Data |
| title_full_unstemmed | S3GRN: Structural Similar Stepwise Generative Recognizable Network for Human Action Recognition With Limited Training Data |
| title_short | S3GRN: Structural Similar Stepwise Generative Recognizable Network for Human Action Recognition With Limited Training Data |
| title_sort | s3grn structural similar stepwise generative recognizable network for human action recognition with limited training data |
| topic | Generative network recognizable network combination structural similar constraint |
| url | https://ieeexplore.ieee.org/document/9272277/ |
| work_keys_str_mv | AT chaoboli s3grnstructuralsimilarstepwisegenerativerecognizablenetworkforhumanactionrecognitionwithlimitedtrainingdata AT xulinshen s3grnstructuralsimilarstepwisegenerativerecognizablenetworkforhumanactionrecognitionwithlimitedtrainingdata AT hongjunli s3grnstructuralsimilarstepwisegenerativerecognizablenetworkforhumanactionrecognitionwithlimitedtrainingdata |