luoww1992
luoww1992
my env: win10 vs2019 libtorch 1.7.1 -cuda101-debug cuda101 i can run it by change the device in torch::Device("cpu"); but how to use it in gpu? How did you do it...
i run it in win10 + linux sucessfully, but i find it is a low rate of utilization in GPU , how to increase it ?
i have see your work, it is beautiful ! i have do all works before training, and i have see the HRNet lib for face landmark, it seem to have...
if --smooth3d in mv1p.py, i can't find the func smooth_skeleton ? where is it ?
there are many wrong in demo with openpose, so do you have a plan to fix it ? and i notice the args --smooth3d has no func to do it.
i notice that you say you can converte to BVH motion frames from video in real-time, and can use it in blender, i want to send bvh mocap data to...
i notice we can get the smpl image with render, how can i get the 3d smpl pose with IK?
now i have a human 3d keypoints file. how can i get the pose euler rotation angle or quaternion with parent-child relationship like smpl pose, i want to use it...
my GPU is 2080ti, when run code, show low gpu utilization, some methods to add it to add fps?
when i use it in smpl with SMPL_MODEL_PATH, this is no some args in pkl model . models.py: 