Hugging Face's logo

Datasets:
Spacewanderer8263
/
Proxy3D-SpaceSpan-318K

Tasks:
Video-Text-to-Text
Modalities:
3D
ArXiv:
Tags:
3d
spatial-intelligence
vlm
visual-grounding
License:
Dataset card Data Studio Files
xet
 Community
1

Add dataset card and link to paper

#1
by nielsr HF Staff - opened
base: refs/heads/main
←
from: refs/pr/1
Discussion Files changed
+44
-3
nielsr

This PR improves the dataset card for the SpaceSpan dataset introduced in the "Proxy3D" paper.

  • Added the video-text-to-text task category.
  • Added links to the paper, project page, and GitHub repository.
  • Included a description of the dataset and its role in the Proxy3D multi-stage training pipeline.
  • Added the BibTeX citation from the official repository.
Add dataset card and link to paper79c8d100
Spacewanderer8263
 Owner

Thanks for reaching out, Niels! These updates look great and really help clarify the model's usage. I've merged the changes. Looking forward to more interactions with the Hugging Face community!
Spacewanderer8263 changed pull request status to merged

Sign up or log in to comment