Skip to main content

Table 1 The collected datasets. The number indicates the volume of data. The test data is an actor’s data(dataset ID’s actor). The training and validation data are comprised of other 6 actors’ mixed data

From: 3D human pose estimation model using location-maps for distorted and disconnected images by a wearable omnidirectional camera

Dataset Test Training Validation
A 18,050 94,713 18,943
B 17,045 95,550 19,111
C 19,754 93,293 18,659
D 17,695 95,009 19,002
E 19,320 93,655 18,731
F 20,627 92,565 18,514
G 19,215 93,742 18,749