Human Activity Recognition in the Presence of Occlusion
The presence of occlusion in human activity recognition (HAR) tasks hinders the performance of recognition algorithms, as it is responsible for the loss of crucial motion data. Although it is intuitive that it may occur in almost any real-life environment, it is often underestimated in most research...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-05-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1424-8220/23/10/4899 |
| _version_ | 1797598339229810688 |
|---|---|
| author | Ioannis Vernikos Theodoros Spyropoulos Evaggelos Spyrou Phivos Mylonas |
| author_facet | Ioannis Vernikos Theodoros Spyropoulos Evaggelos Spyrou Phivos Mylonas |
| author_sort | Ioannis Vernikos |
| collection | DOAJ |
| description | The presence of occlusion in human activity recognition (HAR) tasks hinders the performance of recognition algorithms, as it is responsible for the loss of crucial motion data. Although it is intuitive that it may occur in almost any real-life environment, it is often underestimated in most research works, which tend to rely on datasets that have been collected under ideal conditions, i.e., without any occlusion. In this work, we present an approach that aimed to deal with occlusion in an HAR task. We relied on previous work on HAR and artificially created occluded data samples, assuming that occlusion may prevent the recognition of one or two body parts. The HAR approach we used is based on a Convolutional Neural Network (CNN) that has been trained using 2D representations of 3D skeletal motion. We considered cases in which the network was trained with and without occluded samples and evaluated our approach in single-view, cross-view, and cross-subject cases and using two large scale human motion datasets. Our experimental results indicate that the proposed training strategy is able to provide a significant boost of performance in the presence of occlusion. |
| first_indexed | 2024-03-11T03:20:51Z |
| format | Article |
| id | doaj.art-d92e574136a1465eb92e31009be681a3 |
| institution | Directory Open Access Journal |
| issn | 1424-8220 |
| language | English |
| last_indexed | 2024-03-11T03:20:51Z |
| publishDate | 2023-05-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj.art-d92e574136a1465eb92e31009be681a32023-11-18T03:14:14ZengMDPI AGSensors1424-82202023-05-012310489910.3390/s23104899Human Activity Recognition in the Presence of OcclusionIoannis Vernikos0Theodoros Spyropoulos1Evaggelos Spyrou2Phivos Mylonas3Department of Informatics and Telecommunications, University of Thessaly, 35131 Lamia, GreeceDepartment of Digital Systems, University of Piraeus, 18534 Piraeus, GreeceDepartment of Informatics and Telecommunications, University of Thessaly, 35131 Lamia, GreeceDepartment of Informatics and Computer Engineering, University of West Attica, Egaleo Park, 12243 Athens, GreeceThe presence of occlusion in human activity recognition (HAR) tasks hinders the performance of recognition algorithms, as it is responsible for the loss of crucial motion data. Although it is intuitive that it may occur in almost any real-life environment, it is often underestimated in most research works, which tend to rely on datasets that have been collected under ideal conditions, i.e., without any occlusion. In this work, we present an approach that aimed to deal with occlusion in an HAR task. We relied on previous work on HAR and artificially created occluded data samples, assuming that occlusion may prevent the recognition of one or two body parts. The HAR approach we used is based on a Convolutional Neural Network (CNN) that has been trained using 2D representations of 3D skeletal motion. We considered cases in which the network was trained with and without occluded samples and evaluated our approach in single-view, cross-view, and cross-subject cases and using two large scale human motion datasets. Our experimental results indicate that the proposed training strategy is able to provide a significant boost of performance in the presence of occlusion.https://www.mdpi.com/1424-8220/23/10/4899human activity recognitionocclusiondeep learningconvolutional neural networks |
| spellingShingle | Ioannis Vernikos Theodoros Spyropoulos Evaggelos Spyrou Phivos Mylonas Human Activity Recognition in the Presence of Occlusion Sensors human activity recognition occlusion deep learning convolutional neural networks |
| title | Human Activity Recognition in the Presence of Occlusion |
| title_full | Human Activity Recognition in the Presence of Occlusion |
| title_fullStr | Human Activity Recognition in the Presence of Occlusion |
| title_full_unstemmed | Human Activity Recognition in the Presence of Occlusion |
| title_short | Human Activity Recognition in the Presence of Occlusion |
| title_sort | human activity recognition in the presence of occlusion |
| topic | human activity recognition occlusion deep learning convolutional neural networks |
| url | https://www.mdpi.com/1424-8220/23/10/4899 |
| work_keys_str_mv | AT ioannisvernikos humanactivityrecognitioninthepresenceofocclusion AT theodorosspyropoulos humanactivityrecognitioninthepresenceofocclusion AT evaggelosspyrou humanactivityrecognitioninthepresenceofocclusion AT phivosmylonas humanactivityrecognitioninthepresenceofocclusion |