Details regarding few-shot and zero shot datasets

Question

Details regarding few-shot and zero shot datasets

Closed this issue 9 months ago · 5 comments

Hi,

Thank you for the code and the Readme which are both very well organized. I am trying to setup the few-shot and zero-shot datasets. Is there any details I need to take account of?

Thank you!.

Answer 1 · 2024-04-21T07:02:19.000Z

Both zero-shot and few-shot datasets we use are supported by mmdetection and mmsegmentation. Please refer to the official documentation for downloading. We also provide instructions here.

One thing to note is that annotations for detection datasets need to be converted to COCO format.

As for config files, we have provided details here.

Answer 2 · 2024-04-21T15:59:03.000Z

Thank you for the details!. It helps a lot!

Answer 3 · 2024-04-26T20:10:12.000Z

Hi,

Sorry for the delayed response. Some of the few shot dataset convertor scripts do not seem to be there.
tools/dataset_converters/drive.py
tools/dataset_converters/loveda.py
tools/dataset_converters/potsdam.py

Could you please provide them?

Thank you.

Answer 4 · 2024-04-27T13:16:37.000Z

We will check it in a few days. :) Thanks for your patience.

Answer 5 · 2024-05-04T03:19:29.000Z

We have updated scripts according to mmsegmentation, you can try again now.