Hierarchical Human Action Recognition with Self-Selection Classifiers via Skeleton Data<xref rid="ctp_70_5_633fn1" ref-type="fn">*</xref><fn id="ctp_70_5_633fn1"><label>*</label><p>Supported by the National Nature Science Foundation of China under Grant Nos. 11475003, 61603003, and 11471093; the Key Project of Cultivation of Leading Talents in Universities of Anhui Province under Grant No. gxfxZD2016174; Funds of Integration of Cloud Computing and Big Data; Innovation of Science and Technology of Ministry of Education of China under Grant No. 2017A09116; and Anhui Provincial Department of Education Outstanding Top-Notch Talent-Funded Project under Grant No. gxbjZD26</p></fn>

Hierarchical Human Action Recognition with Self-Selection Classifiers via Skeleton Data**

Supported by the National Nature Science Foundation of China under Grant Nos. 11475003, 61603003, and 11471093; the Key Project of Cultivation of Leading Talents in Universities of Anhui Province under Grant No. gxfxZD2016174; Funds of Integration of Cloud Computing and Big Data; Innovation of Science and Technology of Ministry of Education of China under Grant No. 2017A09116; and Anhui Provincial Department of Education Outstanding Top-Notch Talent-Funded Project under Grant No. gxbjZD26

Su Ben-Yue^{1, 2, †}, Wu Huang², Sheng Min^{2, 3}, Shen Chuan-Sheng^{3, ‡}

The voting mechanism of OCU. Step D1 is to obtain the probability of each classifier having maximal recognition rate in M tests, and outputs the index of the classifier with maximal probability, D2 calculates the average rate over M trials of the classifiers offered by D1, and outputs the index of the classifier with maximal average rate, and D3 randomly chooses a classifier given by D2.