New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

question of bbox in VQA feature #61

Open

1144181135 opened this issue Apr 25, 2021 · 1 comment

1144181135 commented Apr 25, 2021

Hi VilBert team, I read your code of VQA dataset: vilbert/datasets/vqa_dataset.py. And I see the bbox is shape of "N x 5". I have some questions:

what's the meaning of dimension 5? A bbox information maybe [x1, y1, x2, y2] with dimension 4. What's the last dimension mean? Classification score?
Does the bbox coordinations [x1, y1, x2, y2] need to be normalized to 0~1 ?
That's all. Thank you.

jiyt17 commented Jul 14, 2021

Hello bro, the answer can be found in VILBERT paper and /vilbert/datasets/_image_features_reader.py

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment