Prepare

   1. download mvb dataset from http://volumenet.cn/#/
   2.> mkdir MVB
     > mv MVB_train/Image       MVB/MVB_train/Image
     > mv MVB_val/Image/gallery MVB/MVB_val/Image/gallery
     > mv MVB_val/Image/probe   MVB/MVB_val/Image/probe
   3. Install dependencies
       pytorch>=1.0
       torchvision
       yacs
       tensorboardX
   4. pretrained model
       You can download the ImageNet pre-trained weights from here
       https://drive.google.com/drive/folders/1thS2B8UOSBi_cJX6zRy6YYRww_nVFI_S

Train

   > python train.py

Evaluator

   > python evaluator.py

Result

mAP	Rank-1	Rank-5	Rank-10
82.3%	80.4%	90.5%	95.0%