TY - JOUR
TI - An adversarial semi-supervised approach for action recognition from pose information
AU - Pikramenos, G.
AU - Mathe, E.
AU - Vali, E.
AU - Vernikos, I.
AU - Papadakis, A.
AU - Spyrou, E.
AU - Mylonas, P.
JO - Neural Computing and Applications
PY - 2020
VL - 32
TODO - 23
SP - 17181-17195
PB - Springer Science and Business Media Deutschland GmbH
SN - null
TODO - 10.1007/s00521-020-05162-5
TODO - Cameras, Action recognition;  Data synthesizers;  Domain adaptation;  Environmental conditions;  Generalization capability;  Global distribution;  Measurement bias;  Pose information, Classification (of information)
TODO - 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.
ER -