TransT win10+cuda11.0环境配置

相关链接

代码地址：https://github.com/chenxin-dlut/TransT
陈鑫的bili视频讲解地址：https://www.bilibili.com/video/BV1654y157cS
paper地址：https://arxiv.org/abs/2103.15436

环境配置

• 参考GitHub的环境配置教程
  https://github.com/chenxin-dlut/TransT
• 下载git 项目

git clone [email protected]:chenxin-dlut/TransT.git

• 创建conda环境

conda create --name transT python=3.7
conda activate transT
conda install pytorch==1.7.1 torchvision==0.8.2 torchaudio==0.7.2 cudatoolkit=11.0 -c pytorch

conda install matplotlib pandas tqdm
pip install opencv-python tb-nightly visdom scikit-image tikzplotlib gdown
conda install cython scipy

pip install pycocotools jpeg4py
pip install wget yacs
pip install shapely==1.6.4.post2

设置环境变量

# Change directory to 
cd TransT

# Environment settings for pytracking. Saved at pytracking/evaluation/local.py
python -c "from pytracking.evaluation.environment import create_default_local_file; create_default_local_file()"

# Environment settings for ltr. Saved at ltr/admin/local.py
python -c "from ltr.admin.environment import create_default_local_file; create_default_local_file()"

下载模型文件，放在pytracking/networks下即可

运行自己的视频：
脚本：E:\pycharmpro\mysot\TransT\pytracking\run_video.py

parser.add_argument('--tracker_name',default="transt", type=str, help='Name of tracking method.')
parser.add_argument('--tracker_param',default="transt50", type=str, help='Name of parameter file.')
parser.add_argument('--videofile', default=r"your video path",type=str, help='path to a video file.')