Ning-Ding
commented
7 years ago Hi~
I have received many emails to ask for the version that I used for this project. Actually, I could tell the true version, but I think the most efficient way is to update the code to a new version such as TensorFlow 1.1 support. So I have make a plan to update the whole project code to a totally new version using TensorFlow 1.1. If you could wait for two weeks, you will see the new code.
For the CUHK03, the paper said nothing about that. After imshow several images of CUHK03, I found by myself that the index 0-4 and 5-9 are correspond to two cameras.
Regards, Ning
2017-05-01 2:05 GMT+08:00 MarcosPieras notifications@github.com:
Hi, First of all, thanks for this work. I am trying to run both models, but I think that I have some problems with the versions of the libraries. Which versions did you use ?
In CUHK03 dataset, you divide the dataset into 'a' and 'b', according to the views of the cameras, they don't say nothing like this in the paper. I am newbie with this dataset, is it normal ?
Thanks!!!
— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/Ning-Ding/Implementation-CVPR2015-CNN-for-ReID/issues/16, or mute the thread https://github.com/notifications/unsubscribe-auth/AKTknIq_QmYAgoAJcJpcgsy4e4EYsx5hks5r1M17gaJpZM4NMpJS .
Hi, First of all, thanks for this work. I am trying to run both models, but I think that I have some problems with the versions of the libraries. Which versions did you use ?
In CUHK03 dataset, you divide the dataset into 'a' and 'b', according to the views of the cameras, they don't say nothing like this in the paper. I am newbie with this dataset, is it normal ?
Thanks!!!