dc.creator | Pikramenos G., Mathe E., Vali E., Vernikos I., Papadakis A., Spyrou E., Mylonas P. | en |
dc.date.accessioned | 2023-01-31T09:50:07Z | |
dc.date.available | 2023-01-31T09:50:07Z | |
dc.date.issued | 2020 | |
dc.identifier | 10.1007/s00521-020-05162-5 | |
dc.identifier.issn | 09410643 | |
dc.identifier.uri | http://hdl.handle.net/11615/78218 | |
dc.description.abstract | The collection of video data for action recognition is very susceptible to measurement bias; the equipment used, camera angle and environmental conditions are all factors that majorly affect the distribution of the collected dataset. Inevitably, training a classifier that can successfully generalize to new data becomes a very hard problem, since it is impossible to gather general enough training sets. Recent approaches in the literature attempt to solve this problem by augmenting a given training set, with synthetic data, so as to better represent the global distribution of the covariates. However, these approaches are limited because they essentially involve hand-crafted data synthesizers, which are typically hard to implement and problem specific. In this work, we propose a different approach to tackling the above issues, which relies on the combination of two techniques: pose extraction, and domain adaptation as a means to improve the generalization capabilities of classifiers. We show that adapted skeletal representations can be retrieved automatically in a semi-supervised setting and these help to generalize classifiers to new forms of measurement bias. We empirically validate our approach for generalizing across different camera angles. © 2020, Springer-Verlag London Ltd., part of Springer Nature. | en |
dc.language.iso | en | en |
dc.source | Neural Computing and Applications | en |
dc.source.uri | https://www.scopus.com/inward/record.uri?eid=2-s2.0-85087712051&doi=10.1007%2fs00521-020-05162-5&partnerID=40&md5=6d8d7bed59f4627b62def485b43c1377 | |
dc.subject | Cameras | en |
dc.subject | Action recognition | en |
dc.subject | Data synthesizers | en |
dc.subject | Domain adaptation | en |
dc.subject | Environmental conditions | en |
dc.subject | Generalization capability | en |
dc.subject | Global distribution | en |
dc.subject | Measurement bias | en |
dc.subject | Pose information | en |
dc.subject | Classification (of information) | en |
dc.subject | Springer Science and Business Media Deutschland GmbH | en |
dc.title | An adversarial semi-supervised approach for action recognition from pose information | en |
dc.type | journalArticle | en |