View Code? Open in Web Editor
NEW
A pytorch implementation of person keypoint detection task in "OpenPose: Realtime Multi-Person 2D Pose Estimation using Part Affinity Fields" at T-PAMI 2019.
License: MIT License
openpose-pytorch's People
Contributors
Watchers
openpose-pytorch's Issues
for example:
Mconv1_stage0_L2_0: channel output is 96
layer {
name: "Mconv1_stage0_L2_0"
type: "Convolution"
bottom: "conv4_4_CPM"
top: "Mconv1_stage0_L2_0"
convolution_param {
num_output: 96
pad: 1
kernel_size: 3
}
}
but in your network the same channel output is 128:
{'conv_stage%d_denseblock1_1' % i: [num_channel_in, 128, 3, 1, 1]},
could u say something about this?thanks a lot
from lib import transforms, datasets
where is the "lib"?can u release it?