VITA icon indicating copy to clipboard operation
VITA copied to clipboard

Published 20 hours ago • verified publisher icon sukjunhwang

Dateset prepare

Open ChinChyi opened this issue 1 year ago • 1 comments

Why "STEP-2: Prepare annotations for combined data"? Why do we need to use the COCO dataset again for the second round of fine-tuning, even though it has already been used for pre-training?

ChinChyi avatar Apr 27 '23 03:04 ChinChyi

It's due to the sparse number of videos of video datasets. COCO gets augmented into videos, and used for training together.

sukjunhwang avatar May 07 '23 05:05 sukjunhwang