Table 1 The receptive field comparison between V block and the block in HRNet.
From: An efficient and accurate 2D human pose estimation method using VTTransPose network
Architecture | y1 | y2 | y3 | y4 |
|---|---|---|---|---|
HRNet_Bottleneck block | 3 | 3 | 3 | 3 |
HRNet_Basic block | 3, 5 | 3, 5 | 3, 5 | 3, 5 |
V block | 3, 15 | 5, 13 | 7, 11 | 9 |
