Are you sure you want to delete this task? Once this task is deleted, it cannot be recovered.
Jiahong XU 7181860033 | 2 years ago | |
---|---|---|
cam_bn_info-/pytorch-ckpt | 2 years ago | |
frameworks | 2 years ago | |
io_stream | 2 years ago | |
pytorch-ckpt/current | 2 years ago | |
testimgs | 2 years ago | |
utils | 2 years ago | |
.gitattributes | 2 years ago | |
cam_bn_info-market_demo-duke.pkl | 2 years ago | |
cam_bn_info-market_demo-market.pkl | 2 years ago | |
cam_bn_info-market_demo-msmt.pkl | 2 years ago | |
cbn_demo.py | 2 years ago | |
cbn_demo_image2.py | 2 years ago | |
config.py | 2 years ago | |
get-pip.py | 2 years ago | |
path_dict.pkl | 2 years ago | |
paths.pkl | 2 years ago | |
readme.md | 2 years ago | |
requirements.txt | 2 years ago | |
test_model.py | 2 years ago | |
train_model.py | 2 years ago |
以摄像机为基本单位的批归一化技术用于行人图像再识别特征提取
Camera-based Batch Normalization for Person Re-identification
Yongfei Zhang, Tianyu Zhang
This repo has the training and testing codes of supervised learning with Camera-based batch normalization Rethinking the Distribution Gap of Person Re-identification with Camera-based Batch Normalization. It aligns camera distributions of both training cameras and testing cameras.
We design a demo for testing.
The training phase uses about 10G of GPU memory.
The ReID accuarcy can be seen in CBN.
Rank-1 accuracy and mAP.
Market-1501 , DukeMTMC-reID and MSMT17 are used.
torch==1.3.1
torchvision==0.4.2
tensorboard
future
fire
tqdm
Name | Notes |
---|---|
Input | The file path of one person image. |
Output | An array containing the feature vector. |
1. Train
CUDA_DEVICE_ORDER=PCI_BUS_ID CUDA_VISIBLE_DEVICES=0,1 \
python train_model.py train --trainset_name market --save_dir='market_demo' --max_epoch 60 --decay_epoch 40
2. Evaluate
CUDA_DEVICE_ORDER=PCI_BUS_ID CUDA_VISIBLE_DEVICES=0 \
python test_model.py test --testset_name market --save_dir='market_demo'
You may refer to config.py to adjust more hyper-parameters.
The demo code is implemented in cbn_demo.py. We write a new class for users to call.
以摄像机为基本单位的批归一化技术用于行人图像再识别特征提取算法
Pickle Python Text
Dear OpenI User
Thank you for your continuous support to the Openl Qizhi Community AI Collaboration Platform. In order to protect your usage rights and ensure network security, we updated the Openl Qizhi Community AI Collaboration Platform Usage Agreement in January 2024. The updated agreement specifies that users are prohibited from using intranet penetration tools. After you click "Agree and continue", you can continue to use our services. Thank you for your cooperation and understanding.
For more agreement content, please refer to the《Openl Qizhi Community AI Collaboration Platform Usage Agreement》