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

数据集构建 #60

Open
Hriver-J opened this issue Feb 21, 2025 · 0 comments
Open

数据集构建 #60

Hriver-J opened this issue Feb 21, 2025 · 0 comments

Comments

@Hriver-J
Copy link

Hriver-J commented Feb 21, 2025

作者您好,我想要增添一部分数据集分别用于训练和测试,想请问你们是如何构建训练数据集的?
更具体一些,在有RGB与RGB-D图像后我该怎样生成grasp_label、collision_label路径以及其他数据集必须的内容?

————————————————————————————————————————————————

根据对graspNet-1billion论文的阅读,我目前的认知是这样的:

  1. 对物体进行3d建模,并生成点云。这一步可以通过3d建模软件生成.obj系列的文件后,使用pcl库来生成点云。
  2. 进行相机、物体标定,将物体位姿统一到相机坐标系下。
  3. 基于机械臂夹爪的建模(获取夹爪间距、深度),遍历物体的点云来生成抓取位姿,同时进行力封闭分析。
  4. 剔除存在碰撞的抓取。

我对“生成抓取位姿“和“力封闭分析”的实现有些困惑:

  1. “生成抓取位姿”,搜索3d点云中是否有多个点落在同一个球面上,倘若是,则说明夹爪有可行的接触。我不明白该如何将这些同球面点和夹爪位姿对应起来?
  2. “力封闭分析”获取抓取评分(摩擦因数)的逻辑该如何实现?我无法将3d点云和力学分析联系起来,不知道您是否有相关的代码实现或者相关的书籍/paper/开源代码分享?
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

1 participant