安装启动lama-cleaner教程

Dear 丶 2024-03-24 23:19 49阅读 0赞

#### 一、下载lama-cleaner到本地 ####

[https://github.com/Sanster/lama-cleaner][https_github.com_Sanster_lama-cleaner]

![d0cc86f66b494ff5ac3646ab6d9f5bcb.png][]

#### 二、安装miniconda ####

参考：[安装启动yolo5教程\_苍穹之跃的博客-CSDN博客][yolo5_-CSDN]

#### 三、创建虚拟环境 ####

![93e4a2003cb349d7966166b63deae63b.png][]

conda create -n lama python=3.8

conda activate lama

进入项目根目录

![6b0f38b9e5ec4671884c1c275e6d1e65.png][]

先安装opencv-python

pip3 install opencv-python -i https://pypi.tuna.tsinghua.edu.cn/simple

安装所有依赖

pip install -r requirements.txt  -i https://pypi.tuna.tsinghua.edu.cn/simple

安装显卡驱动：[安装启动yolo5教程\_苍穹之跃的博客-CSDN博客][yolo5_-CSDN]

安装pytorch

conda install pytorch==1.10.0 torchvision==0.11.0  cudatoolkit=11.3 -c pytorch -c conda-forge

#### 四、启动项目 ####

python main.py --model=lama --device=cpu --port=8080

![fc751db558d84cad97deb94eb7d780b8.png][]

此时系统自动会去远程下载模型（大概会失败！！！）

![0bc158f7e8d6443eaa79c4b8a7166d02.png][]

此时打开梯子，然后再重新启动即可。

![2a224da268914f61b2cc352749f6302f.png][]

![ecc66bc2af20454f84220be69df283d6.png][]

[https_github.com_Sanster_lama-cleaner]: https://github.com/Sanster/lama-cleaner
[d0cc86f66b494ff5ac3646ab6d9f5bcb.png]: https://image.dandelioncloud.cn/pgy_files/images/2024/03/13/ebfb90418dc24c5d923dc5005608b068.png
[yolo5_-CSDN]: https://blog.csdn.net/wenxingchen/article/details/132498194
[93e4a2003cb349d7966166b63deae63b.png]: https://image.dandelioncloud.cn/pgy_files/images/2024/03/13/98dc7bd4b6d64eb2967acd516784e7ed.png
[6b0f38b9e5ec4671884c1c275e6d1e65.png]: https://image.dandelioncloud.cn/pgy_files/images/2024/03/13/ee130c4e140e4e108ecda44baf14dc05.png
[fc751db558d84cad97deb94eb7d780b8.png]: https://image.dandelioncloud.cn/pgy_files/images/2024/03/13/5bfb791415e441c2a9da815172ffc68b.png
[0bc158f7e8d6443eaa79c4b8a7166d02.png]: https://image.dandelioncloud.cn/pgy_files/images/2024/03/13/4292acad7552493398b5a9af9bee9836.png
[2a224da268914f61b2cc352749f6302f.png]: https://image.dandelioncloud.cn/pgy_files/images/2024/03/13/73c07e71c4204e70879b82ff677fc998.png
[ecc66bc2af20454f84220be69df283d6.png]: https://image.dandelioncloud.cn/pgy_files/images/2024/03/13/2e4e69d23af7473e85002ac6c04cc458.png