caa a22 4500 605446849 CHVBK 20210128100129.0 cr unu---uuuuu 210128e20150101xx s 000 0 eng 10.1007/s11042-014-1936-z doi (NATIONALLICENCE)springer-10.1007/s11042-014-1936-z Evaluation of semi-supervised learning method on action recognition [Elektronische Daten] [Haoquan Shen, Yan Yan, Shicheng Xu, Nicolas Ballas, Wenzhi Chen] Action recognition is one of the most difficult problems in computer vision and multimedia areas, since both spatial information and spatiotemporal semantic meaning should be taken into consideration. Moreover, the noisy and weakly annotated information make this task even harder. Nowadays, instead of the traditional features and classifiers, a lot of new attempts have made the task of action recognition promising. Noticing that there is no work on comparison of different combination of pooling and semi-supervised learning method under the same experiment setting, it would be interesting to apply different combination of pooling and semi-supervised learning method on both the synthetic and realistic action recognition datasets to see which combination or method performs better. In summary, we can obtain the following conclusions based on our experiments. Firstly, Second Order Pooling (Carreira et al.2012) is worse than the traditional Bag of Words (Schmid and Mohr1997; Dance et al.2004) regarding to the overall performance in some dataset, but is a good way to speed up the coding stage of video classification with little sacrifice of performance. Secondly, Semi-supervised Hierarchical Regression Algorithm (MLHR) and Manifold Regularized Least Square Regression (MRLS) (Belkin et al. J Mach Learn Res 12:2399-2434,2006) is better than some of the supervised learning methods (χ 2-SVM, SVM-2K (Farquhar et al.2006)) in the real world action recognition problems which shares little available annotated information. Thirdly, for KTH, UCF50 and HMDB dataset, late fusion doesn't necessarily improve the performance. In comparison, MLHR, SVM-2K and Multi-kernel Learning is a more natural way to deal with multi-feature problems. Springer Science+Business Media New York, 2014 Semi-supervised learning nationallicence Multi-feature fusion nationallicence Second order pooling nationallicence Video concept annotation nationallicence Action recognition nationallicence Shen Haoquan Department of Computer Science, Zhejiang University, Zhejiang, China aut Yan Yan Department of Information Engineering and Computer Science, University of Trento, Trento, Italy aut Xu Shicheng Department of Computer Science, Zhejiang University, Zhejiang, China aut Ballas Nicolas CEA and Mines-ParisTech, Paris, France aut Chen Wenzhi Department of Computer Science, Zhejiang University, Zhejiang, China aut Multimedia Tools and Applications Springer US; http://www.springer-ny.com 74/2(2015-01-01), 523-542 1380-7501 74:2<523 2015 74 11042 https://doi.org/10.1007/s11042-014-1936-z text/html Onlinezugriff via DOI BK010053 XK010053 XK010000 Metadata rights reserved Springer special CC-BY-NC licence nationallicence 1 research-article jats NATIONALLICENCE NATIONALLICENCE NL-springer NATIONALLICENCE 856 40 https://doi.org/10.1007/s11042-014-1936-z text/html Onlinezugriff via DOI NATIONALLICENCE 700 1- Shen Haoquan Department of Computer Science, Zhejiang University, Zhejiang, China aut NATIONALLICENCE 700 1- Yan Yan Department of Information Engineering and Computer Science, University of Trento, Trento, Italy aut NATIONALLICENCE 700 1- Xu Shicheng Department of Computer Science, Zhejiang University, Zhejiang, China aut NATIONALLICENCE 700 1- Ballas Nicolas CEA and Mines-ParisTech, Paris, France aut NATIONALLICENCE 700 1- Chen Wenzhi Department of Computer Science, Zhejiang University, Zhejiang, China aut NATIONALLICENCE 773 0- Multimedia Tools and Applications Springer US; http://www.springer-ny.com 74/2(2015-01-01), 523-542 1380-7501 74:2<523 2015 74 11042