Skip to content
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.

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

How are region-level descriptions obtained? #16

Open
UcanSee opened this issue Apr 22, 2024 · 1 comment
Open

How are region-level descriptions obtained? #16

UcanSee opened this issue Apr 22, 2024 · 1 comment

Comments

@UcanSee
Copy link

UcanSee commented Apr 22, 2024

Thanks for your great work!
In your paper, the label of the semantic classification branch is mask cropped embedding obtained by CLIP, then how is the GT of the caption branch generated from SA-1B?

@PhyscalX
Copy link
Collaborator

PhyscalX commented Apr 22, 2024

Hi, @UcanSee

  1. Caption branch (i.e., TextDecoder) is randomly initialized, but is NOT trained during SA-1B pre-training.

  2. Caption branch is then trained only on VG data, with the frozen ImageEncoder & ImageDecoder.

  3. Further e2e fine-tuning for Caption branch on a mixed SA/VG dataset (set caption loss to zero for SA data),
    could improve VG CIDEr from 154.7 to 164.7.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants