Table 5 VisionTool’s detection accuracy on MOCA dataset. A k-fold (k \(=\) 5) approach is used for each view point (i.e., the detectors are trained on fourfolds and the remaining one was predicted). The results reported in the table correspond to the average mAP computed across the different folds.
From: A semi-automatic toolbox for markerless effective semantic feature extraction
View point | mAP\(^{0.5}\) | mAP\(^{0.75}\) | mAP\(_{\text {index}}\) | mAP\(_{\text {little finger}}\) | mAP\(_{\text {hand}}\) | mAP\(_{\text {wrist}}\) | mAP\(_{\text {elbow}}\) | mAP |
|---|---|---|---|---|---|---|---|---|
Lateral | 0.969 | 0.905 | 0.865 | 0.845 | 0.889 | 0.958 | 0.988 | 0.909 |
Egocentric | 0.962 | 0.929 | 0.925 | 0.789 | 0.963 | 0.922 | 0.978 | 0.915 |
Frontal | 0.957 | 0.858 | 0.861 | 0.907 | 0.836 | 0.930 | 0.992 | 0.905 |
All together | 0.954 | 0.904 | 0.880 | 0.821 | 0.912 | 0.949 | 0.980 | 0.908 |
