MMEditing is a low-level vision toolbox based on PyTorch, supporting super-resolution, inpainting, matting, video interpolation, etc.
computer-vision
deep-learning
pytorch
generative-adversarial-network
image-generation
super-resolution
inpainting
matting
video-super-resolution
video-interpolation
video-frame-interpolation
-
Updated
Apr 7, 2022 - Python
There is a feature that allows to use donor areas for the inpainting process. How does it works? You can provide an additional mask image where masked areas that fall into the area to inpaint will be inpainted from the areas that are not falling into the inpaint area.
The task is pretty fun though.