A Pytorch Implementation of "Unlabeled Samples Generated by GAN Improve the Person Re-identification Baseline in vitro"(ICCV17)
158
stars
49
forks
source link
Why are the experimental results much higher than the results in the paper? #24
Open
XXXxiaoxiaoxia opened 5 years ago
Why are the experimental results much higher than the results in the paper?
What is the difference between the loss function and the network structure in the code and the paper mentioned in the paper?