---

If you find this work useful, please kindly cite our paper as,

@article{zheng2019attributes,									   			
  title={Attributes Guided Feature Learning for Vehicle Re-identification},                                    
  author={Zheng, Aihua and Lin, Xianmin and Li, Chenglong and He, Ran and Tang, Jin},                          
  journal={arXiv preprint arXiv:1905.08997},                                                                   
  year={2019}                                                                                                  
}  

This code should be used for research only. Please DO NOT distribute or use it for commercial purpose.

our paper:https://arxiv.org/abs/1905.08997

If you have any problem, please contact xmlin1995@gmail.com

---

Training a Model for Vehicle Re-Identification

In order to train our models for Vehicle Re-Identification, we start of with an Imagenet pre-trained model and fine tune it on our task. These pre-trained models can be found on the Tensorflow Models Readme Page.

For training a Vehicle Re-Id model, the trainer_preid.py script is used. To get a look at all possible arguments, have a look at the script's main method. Here, we'll only show the most important ones.

Training a Resnet-50 Baseline on VeRi-776

To run our training script for a Resnet-50 Baseline model with VeRi-776, execute the following while replacing all the <> tags with the corresponding values.

python3 trainer_preid.py --output=<output directory> --data=<dataset directory> --dataset-name=Veid_view --batch-size=16 --num-epochs=100 --network-name=resnet_v1_50 --initial-checkpoint=<path to imagenet checkpoint or another checkpoint you want to load> --checkpoint-exclude-scopes=resnet_v1_50/logits --trainable-scopes=resnet_v1_50/logits --no-evaluation

Training the Views Predictor

In order to train our Views model, you need to have a dataset providing views information. In our paper, we used VeRi-776 for training the view predictor, so we label the view information by ourself which can be find on http://www.escience.cn/people/AihuaZheng/Code-Dataset.html. The trainer_views.py script can be called as follows:

python3 trainer_views.py --output=<output directory> --data=<dataset directory> --dataset-name=Veid_view --batch-size=16 --num-epochs=100 --network-name=resnet_v1_50_views --initial-checkpoint=<path to imagenet checkpoint or another checkpoint you want to load> --checkpoint-exclude-scopes=resnet_v1_50/logits --trainable-scopes=resnet_v1_50/3Views

Training the Types Predictor and Colors Predictor

The training script of Types and Colors are the same as Views Predictor.

Using Tensorboard to track training

During training, you can keep track of the loss and other important numbers by starting Tensorboard.

tensorboard --logdir=<output directory of the training or a parent folder of it>

The results can be viewed by opening a browser and go to localhost:6006.

Feature Prediction

To predict the features, run the predictor_preid.py script:

python3 predictor_preid.py --model-dir=<the model to be loaded> --data=<dataset directory> --dataset-name=Veid --batch-size=128 --network-name=resnet_v1_50_views

The predicted features will be stored in a subfolder of the specified model-dir called predictions.