Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Some questions about human36m_diffpose_uvxyz_cpn.yml #3

Open
jiejie22333 opened this issue May 25, 2023 · 4 comments
Open

Some questions about human36m_diffpose_uvxyz_cpn.yml #3

jiejie22333 opened this issue May 25, 2023 · 4 comments

Comments

@jiejie22333
Copy link

I am new to 3D HPE,may I ask you some questions?
in human36m_diffpose_uvxyz_cpn.yml
data:
dataset: "human36m"
dataset_path: "./data/data_3d_h36m.npz"
##why use gt_gmm?##
dataset_path_train_2d: "./data/data_2d_h36m_gt_gmm.npz"
dataset_path_test_2d: "./data/data_2d_h36m_cpn_ft_h36m_dbb_gmm.npz"
num_joints: 17
num_workers: 32

why path_train uses data_2d_h36m_gt_gmm.npz and path_test uses data_2d_h36m_cpn_ft_h36m_dbb_gmm.npz

@GoogQing
Copy link
Collaborator

Human3,.6m has two settings: use the GT 2D pose or use the CPN detected 2D pose.
During model training, we find that using the GT 2D pose (in the training set) to optimize our diffusion model can achieve better performance on both GT 2D pose (in the test set) and the CPN-detected 2D pose (in the test set). Thus, we use the GT pose for model training in the CPN pose setting.

@jiejie22333
Copy link
Author

thank you !!!

@jiejie22333
Copy link
Author

May I ask when will you release your codes about Video-based experiments ?

@GoogQing
Copy link
Collaborator

GoogQing commented Jun 2, 2023

Sorry for the delay. We are very busy in recent days due to several emergent projects. We will try to release the video-based code by the end of this month.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants