Datasets:

---

jesbu1

/

peek_bridge_labels

like 1

This PR significantly improves the dataset card for the peek_bridge_labels dataset.

Key enhancements include:

• Added task_categories: ['robotics', 'keypoint-detection'] and descriptive tags (robot-manipulation, vision-language-models, zero-shot-generalization, bridge-v2) to the YAML metadata for better organization and searchability on the Hugging Face Hub.
• Included a clear description of the dataset, explaining its content (PEEK VLM path and mask labels for BRIDGE_v2) and its role in enhancing zero-shot generalization for robot manipulation policies.
• Provided direct links to the associated research paper (PEEK: Guiding and Minimal Image Representations for Zero-Shot Generalization of Robot Manipulation Policies), the official project page (https://peek-robot.github.io), and the main PEEK Github repository (https://github.com/peek-robot/peek).
• Added a section describing the sample usage and intent of these labels within the PEEK framework.
• Included the BibTeX citation for proper academic referencing.

These additions make the dataset card more informative and align it with best practices for dataset documentation on the Hugging Face Hub.