Please use the .NET version of waifu2x-nvger from now on.
The ported version includes several additional features and bug fixes.
You can continue to use this version.
However, we will not be able to provide support in case of any problems.
GUI version of the image upscaling tool (waifu2x-ncnn-vulkan).
Provides user interface functionality to waifu2x-ncnn-vulkan (command line).
This tool can upscale not only images but also videos.
The following redistribution package must be installed in order to run this application.
If you have not installed it yet, please download it from above and install it on your PC.
If you do not want to install the redistribution package, you can start the program by manually placing the following DLLs in C:/Windows/System32/ or in the same directory as the EXE.
At least MSVCP140.DLL
, VCRUNTIME140.DLL
, and MFC140U.DLL
are required.
(*Not required for v1.3 or earlier versions)
As a reminder, waifu2x uses the GPU to do the upscaling. Although processing is possible with the built-in graphics, using the GPU allows for faster conversion.
If you want to convert a video, you need to split the frame before upscaling, which takes a lot of time.
You can enlarge in while keeping the image clean.
Examples:
※Click on the image to enlarge.
Original (640x800)
waifu2x (640x800, Reduction Level2, CUnet)
Enlarged (1280x1600, Bilinear interpolation)
Enlarged (1280x1600, Lanczos3 interpolation)
waifu2x (1280x1600, Reduction Level2, CUnet)
waifu2x (1280x1600, Reduction Level2, RGB)
waifu2x (1280x1600, Reduction Level2, Photo)
waifu2x (1280x1600, Reduction Level3, CUnet, TTA)
Image Character: THE IDOLM@STER: Haruka Amami
Image upscaling is the main focus, but video upscaling is also possible.
See the comparison video below.
When upscaling video, first use FFmpeg to perform frame splitting.
When you do this, it puts a very large load on your PC.
If you have a low-spec PC, it may take a very long time.
Basically, the time taken for upscaling is proportional to the time and resolution of the video that is the target of upscaling.
This means that upscaling a video with a long duration and relatively high resolution will take a long time.
However, if your PC is equipped with a powerful GPU, this process can be shortened.
Upscaling speed is highly dependent on the performance of the GPU.
The higher the performance of the GPU, the faster the upscaling speed will be.
We do not recommend using the CPU's built-in graphics (iGPU), but that does not mean it cannot be converted.
However, if you use the built-in graphics, it may take a lot of time.
For this reason, NVIDIA or AMD (RADEON) GPUs are recommended.
OS: Windows 7 or higher 64bit PC.
There is no application for 32-bit (x86).
If you need a 32-bit (x86) version, please build it yourself from the C++ source.
OS: Windows10 64bit
CPU: Intel Core i3 or adove, AMD Ryzen 3 or above
RAM: 8GB or above
GPU: NVIDIA GeForce GTX 1060 or above
Development: Microsoft Visual Studio 2019
OS: Windows10 Pro 64bit, 20H2
CPU: Intel(R) Core(TM) i9-9900K CPU @ 3.60GHz
GPU: NVIDIA Geforce GTX 1070Ti
RAM: 32GB
Used tools
OpenCV
FFmpeg
waifu2x-ncnn-vulkan
This tool is released under the MIT license.