Table 1 Confusion matrix for the UGOKI3D dataset
From: Flexible human action recognition in depth video sequences using masked joint trajectories
(a) | (b) | (c) | (d) | (e) | (f) | (g) | (h) | |
|---|---|---|---|---|---|---|---|---|
(a) | 100 % (9/9) | |||||||
(b) | 88.89 % (8/9) | 11.11 % (1/9) | ||||||
(c) | 88.89 % (8/9) | 11.11 % (1/9) | ||||||
(d) | 100 % (9/9) | |||||||
(e) | 100 % (9/9) | |||||||
(f) | 11.11 % (1/9) | 11.11 % (1/9) | 77.78 % (7/9) | |||||
(g) | 100 % (9/9) | |||||||
(h) | 100 % (9/9) |