Our new X account is live! Follow @wizwand_team for updates
WorkDL logo mark

Skeleton-based Action Recognition on NTU (Cross-Subject)

89.4Accuracy

Ours(Joint+Bone)

36.56850.2846477.716Mar 13, 2018Jun 22, 2018Oct 1, 2018Jan 11, 2019Apr 22, 2019Aug 1, 2019Nov 11, 2019
Updated 4d ago

Evaluation Results

MethodLinks
89.4
2019.11
88.5
2018.11
87.5
2019.11
87.5
2018.11
86.9
2019.11
86.9
2019.11
86.8
2018.11
85
2018.05
84.8
2018.11
84.8
2019.11
84.8
2018.11
84.3
2018.11
83.5
2019.11
83.5
2018.11
83.2
2018.11
82.9
2018.03
81.8
2018.05
81.5
2018.11
81.5
2019.11
81.5
2018.07
80.7
2019.09
80.7
80.03
2019.11
80
2018.07
79.8
2019.09
79.8
79.6
79.6
79.57
2018.07
79.4
2018.05
79.4
2019.09
79.4
2019.11
79.2
2018.03
78.58
77.1
2018.03
76.2
76
75.94
2018.07
75.2
75.2
2018.11
74.9
2019.11
74.9
2018.07
74.6
74.6
2019.09
74.6
2018.07
74.4
2019.09
74.4
2018.03
74.3
2019.11
74.3
2018.07
73.4
2018.05
73.4
2018.03
73.4
2018.03
73.4
2019.11
73.4
2019.09
73.4
71.3
2018.11
70.3
2018.03
70.26
69.2
69.2
69.2
69.2
69.2
2018.03
62.93
62.9
62.9
2018.11
62.9
2019.11
62.9
62.9
2018.03
62.05
61.37
60.2
2018.11
60.2
2019.11
60.2
60.2
2018.03
60.09
2018.03
59.14
2018.07
59.1
2018.05
59.1
2019.09
59.1
2018.03
56.29
2018.03
56.02
2018.07
50.1
2019.09
50.1
38.6
38.6