CVPR2021跟踪算法STMTrack的配置（Template-free Visual Tracking with Space-time Memory Networks）

1、论文下载地址

STMTrack: Template-free Visual Tracking with Space-time Memory Networks CVPR (2021). [paper][code]

2、代码下载地址

https://github.com/fzh0917/STMTrack

3、创造虚拟环境并激活

conda create -n STMTrack python=3.7 -y
source activate STMTrack  

4、安装torch和torchvision

pip install torch===1.4.0 -f https://download.pytorch.org/whl/torch_stable.html

pip install torchvision===0.5.0 -f https://download.pytorch.org/whl/torch_stable.html

5、安装依赖库

pip install -r requirements.txt

 6、预训练模型下载

链接：https://pan.baidu.com/s/16aqJy4eMFTDjEXuWFxvlLg 
提取码：z4aj 

将下载的预训练模型放入工程目录下新建的pretrain_model路径中

7、实验设置

打开~/STMTrack-main/experiments/stmtrack/test/otb/stmtrack-googlenet-otb.yaml

1）更改预训练模型所在路径，即6步骤我们存放的路径

2）更改device_num

3）添加数据集所在路径

data_root: "/home1/publicData/Datasets/OTB100"

8、运行代码

 python main/test.py --config experiments/stmtrack/test/otb/stmtrack-googlenet-otb.yaml

可能遇到如下错误

File "pycocotools/_mask.pyx", line 1, in init pycocotools._mask
ValueError: numpy.ndarray size changed, may indicate binary incompatibility. Expected 88 from C header, got 80 from PyObject

解决方法

numpy版本过低导致，需要对numpy版本进行升级即可！

pip install --upgrade numpy

9、运行成功