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 to reproduce on differnet dataset (e.g. CityFlow)? #7

Open
hurueilin opened this issue Sep 9, 2021 · 3 comments
Open

How to reproduce on differnet dataset (e.g. CityFlow)? #7

hurueilin opened this issue Sep 9, 2021 · 3 comments

Comments

@hurueilin
Copy link

Hi, I saw that your project "is only supported on VeRi-776 dataset currently." But in your paper, you've test mAP on CityFlow dataset. What should I do in order to reproduce on CityFlow dataset?
Thanks for helping!

@tsaishien-chen
Copy link
Owner

To implement on CityFlow dataset, we manually annotate the viewpoint labels for around 8000 images on CityFlow and finetune SPAN that is pre-trained on VeRi-776 dataset.

@hurueilin
Copy link
Author

So you collect 8000 images as training set for CityFlow, am I correct?
For (b) Part Feature Extraction and (c) CPDM, you mentioned this repo. Is the code here exactly the one you used, or you implement (b) and (c) by yourself?
Thanks for your kindly reply!

@tsaishien-chen
Copy link
Owner

Yes, correct.
We implemented them ourselves but with an older baseline code.
As in the paper, our baseline got 59.1% mAP while theirs got 77.2% mAP on VeRi-776.

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