Lossy compression.
The opposite is largely just a movie thing. Only in the last few years has machine learning-based image interpolation become a practical thing, but it’s just *interpolation*, basically an educated guess. It’s reasonably good at sharpening some contours on upscaled images or guessing a generic replacement for a missing part of an image from the surroundings, but it’s not some magic “enhance” that can bring back data lost in compression.
Latest Answers