megvii-research / Iter-E2EDET

Official implementation of the paper "Progressive End-to-End Object Detection in Crowded Scenes"

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Can I use coco format annotation directly?

hoya-cho opened this issue · comments

The format of crowdhuman data is known as the odgt format.
Are you using the crowdhuman data format converted to coco format?
Or do you ignore coordinate values or attribute values that are not in coco format?

and Do I have to input both coco format json(train,eval) and odgt format json(train,eval) into config.py?

commented

We use coco format for both training and testing.