Skip to content
This repository has been archived by the owner on Jan 29, 2021. It is now read-only.

trainging sets #3

Open
lucheng07082221 opened this issue Dec 7, 2017 · 1 comment
Open

trainging sets #3

lucheng07082221 opened this issue Dec 7, 2017 · 1 comment

Comments

@lucheng07082221
Copy link

I want know what is your training sets?is VOC2007?thanks!

@YunqiuXu
Copy link
Owner

YunqiuXu commented Dec 7, 2017

Hi, the data can be collected from google image with the crawler I mentioned in Tutorial.pdf, due to security reasons larger "ISIS handsign" dataset can not be provided. We also use the dataset of Viva Challenge (http://cvrr.ucsd.edu/vivachallenge/index.php/hands/hand-detection/) to pretrain the model, then fine-tune to our dataset. We do not use VOC2007 but all the datasets need to be transformed to that format.

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

No branches or pull requests

2 participants