Datasets:

Languages:
English
ArXiv:
License:

Request for download of sub-volume archives

#24
by limingcv - opened

Hi @mattdeitke ,

Thanks for your excellent contribution, but it is not easy for people to download such a huge dataset. I tried many times to download Objaverse but failed due to the network or something else.

But when I download the CO3D dataset, it was pretty successful. Is it possible to compress the data set into volumes like CO3D? I think this would be more helpful if you can refer to the download steps as the CO3D code.

Hi~@limingcv ,

I want to know if I want to download the full version of Objaverse, where can I find the corresponding download script? Or can you share the script you use to download?

Thanks!

Hi~@limingcv ,

I want to know if I want to download the full version of Objaverse, where can I find the corresponding download script? Or can you share the script you use to download?

Thanks!

HI @xiaoxin666 ,

Here are my code:

import objaverse
import multiprocessing


uids = objaverse.load_uids()
processes = multiprocessing.cpu_count()

objects = objaverse.load_objects(
    uids=uids,
    download_processes=processes
)

lvis_annotations = objaverse.load_lvis_annotations()

Sign up or log in to comment