Forums

The 3rd Large-scale Video Object Segmentation - Track 3: Referring Video Object Segmentation Forum

Go back to competition Back to thread list Post in this thread

> How to split the first frame and full video annotations in meta_expressions.json file

I do not know how to split the first frame and full video annotations in the meta_expressions.json file. Or should I directly use all annotations to train my model?

Posted by: zhaowangbo @ April 29, 2021, 2:20 p.m.

The dataset provided in this challenge only contains full-video expressions, so you can directly use all annotations to train your model.

Posted by: seonguk @ April 30, 2021, 3:54 a.m.
Post in this thread