Skip to content

Tikson是一款开源的图像处理APP,目前包含三大核心功能“LaMa重绘、智能抠图高清放大”,基于NCNN框架,实现多种模型的纯端侧推理。

License

Notifications You must be signed in to change notification settings

Fischer-pixel/Tikson

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

8 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

This image was generated by Flux, fabulous!

Tikson

Tikson is an open source image processing APP, currently contains three core functions "LaMa redrawing, intelligent matting, HD amplification", based on the NCNN framework, to achieve a variety of models of pure end-side inference.

Tikson是一款开源的图像处理APP,目前包含三大核心功能“LaMa重绘、智能抠图、高清放大”,基于NCNN框架,实现多种模型的纯端侧推理。


Download & Build status

https://github.com/Fischer-pixel/Tikson/releases/latest

If you don't want to build from source, from the network location Baidu Netdisk(link: https://pan.baidu.com/s/149u__tMBz-DBGQxDgWZ5Vg?pwd=b3jt extraction code: b3jt) download the full amount of the APP (700 MB).

如果你不想从源代码构建Android项目,可以从[百度网盘](链接: https://pan.baidu.com/s/149u__tMBz-DBGQxDgWZ5Vg?pwd=b3jt 提取码: b3jt)下载全量APP(700MB)。

  1. Download all model to "app/src/main/assets" directory from release.

    release中下载所有模型到"app/src/main/assets"目录中。

  2. Download NCNN-vulkan, and extract it to "app/src/main/cpp" directory.

    下载ncnn-vulkan并解压到"app/src/main/cpp"目录下。

  3. Download the opencv-mobile, and extract it into "app/src/main/cpp" directory.

    下载opencv-mobile并解压到"app/src/main/cpp"目录下。

  4. Change the opencv-mobile and NCNN version num to your own download version in "app/src/main/cpp/CMakeLists.txt".

    更改"app/src/main/cpp/CMakeLists.txt"中opencv-mobile,ncnn为你自己下载的版本号。 cmakelists.txt


References

  1. ncnn
  2. opencv-mobile
  3. Doodle
  4. Real-ESRGAN
  5. Real-ESRGAN-ncnn-vulkan
  6. rembg
  7. u2net
  8. lama
  9. iopaint
  10. pytorch

to be continued... 未完待续。。。

About

Tikson是一款开源的图像处理APP,目前包含三大核心功能“LaMa重绘、智能抠图高清放大”,基于NCNN框架,实现多种模型的纯端侧推理。

Resources

License

Stars

Watchers

Forks

Packages

No packages published