High-quality Alignment Subset Object Ids
#4
by
zhizdev
- opened
Thanks for putting together this awesome dataset!
I am interested in training with the 1.3M subset of Objaverse XL with the high quality alignment mentioned in the paper.
Could object ids of the alignment subset be provided?
Thanks!
yeah, will share soon!
Awesome! Looking forward to it! Thanks :)
Seems like you can now download them directly. There is a example in their Goolge Colab: https://colab.research.google.com/drive/15XpZMjrHXuky0IgBbXcsUtb_0g-XWYmN?usp=sharing