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

数据集中使用的'object_euler_xy'和 'object_init_z'是怎么生成的 #6

Open
guoqingMeng opened this issue Jun 26, 2024 · 4 comments

Comments

@guoqingMeng
Copy link

keys_to_convert = ['target_qpos', 'target_hand_pos', 'target_hand_rot', 'object_euler_xy', 'object_init_z']
我使用DexGraspNet生成的数据集与UniDexGrasp2中的数据集不一致,请问哪里有资料可以查看。

@wkwan7
Copy link
Contributor

wkwan7 commented Jun 26, 2024

请参考DexGraspNet的table branch:https://github.com/PKU-EPIC/DexGraspNet/tree/table

@guoqingMeng
Copy link
Author

我看了table分支,还有几点没搞明白。

  1. table分支生成的物体姿态的维度4x4,怎么和 'object_euler_xy', 'object_init_z'进行转换?
  2. pc_feat是在哪里生成的?

@guoqingMeng
Copy link
Author

plane = data_dict['plane'] # plane parameters (A, B, C, D), Ax + By + Cz + D >= 0, A^2 + B^2 + C^2 = 1 translation, euler = plane2euler(plane, axes='sxyz') # object object_euler_xy = torch.tensor([euler[0], euler[1]], dtype=torch.float, device=self.device) object_init_z = torch.tensor([translation[2]], dtype=torch.float, device=self.device)
shadow_hand_random_load_vision.py 里面这段应该是转换 'object_euler_xy', 'object_init_z'的方式;pc_feat没有看到在那里生成的

@wkwan7
Copy link
Contributor

wkwan7 commented Jun 29, 2024

Hi!pc_feat对应论文中pre-train的一个对物体进行classification的模型输出的每个物体的feature,这部分训练代码我们没有release,目前仅将结果保存release了。

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