snap-research / mmvid Goto Github PK

Thanks for the awesome project !
I cannot help running the codes, and wonder when this codes are released.

Thanks for your inspiring article.
When I test the text_wo_video task, why does the program still need to read the video folder？
Can text_to_video task generate videos from text alone without visual control?
Did I have a problem with the parameters or I misunderstand the task?
Looking forward to your reply @alanspike

Hi,

Thanks for sharing your great work. Can you please elaborate on how you generated the texts for dependent multimodal generation? Are you guys using a fixed template as MUG or Shapes? I might miss the details in your main work or supp, so it would be much appreciated if you can address my concern.

Hi there @phymhan,
Thx for sharing the code. I just found that the link provided for generating texts is broken. Hope it can be fixed.

May I know what is the best way to download the training data? It is very huge (~270G). I tried gdown, but it tells that
Access denied with the following error:

Too many users have viewed or downloaded this file recently. Please
try accessing the file again later. If the file you are trying to
access is particularly large or is shared with many people, it may
take up to 24 hours to be able to view or download the file. If you
still can't access a file after 24 hours, contact your domain
administrator.

Is there any suggestion? Thanks.

amazing work, can you please add a colab to test this.

The whole dataset processed is so huge. Could you please offer a demo or test data image dir?