Hello,
I am working on images with anisotropic pixels.
The inpainting methods are using the gradient to implement inpainting so the pixel size is important. What is the best way to consider anisotropic pixels? resampling the image to have an isometric pixel or convolving the image with a low pass Gaussian filter with updated sigma (dividing by pixel size in x and y)?
Thanks
Hello,
I am working on images with anisotropic pixels.
The inpainting methods are using the gradient to implement inpainting so the pixel size is important. What is the best way to consider anisotropic pixels? resampling the image to have an isometric pixel or convolving the image with a low pass Gaussian filter with updated sigma (dividing by pixel size in x and y)?
Thanks