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

Why are models trained on RLBench data generally suited for Keyframe Position Prediction? #262

Open
fengxiuyaun opened this issue Dec 16, 2024 · 0 comments

Comments

@fengxiuyaun
Copy link

Here are three prediction methods:

  1. Initial Frame Grasp Prediction:

    • This method predicts the grasp point of the target object by analyzing the initial frame.
  2. Keyframe Position Prediction:

    • This approach involves predicting the next keyframe position by inputting a series of keyframes, suitable for tasks that require consideration of intermediate state changes.
  3. Frame-by-Frame Action Prediction:

    • In this method, each frame is inputted, and the next position is outputted, making it ideal for tasks that require continuous monitoring and real-time decision-making.

Why are models trained on RLBench data generally suited for Keyframe Position Prediction?

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