Finding duplicates based on name, size and hash are techniques that have helped users get the job done, but they also have their own limitations. For instance, the hash technique will only detect files that are 100% identical in nature, and even the slightest of changes to a file will cause it to be left out of the duplicates. And filename and size don’t even guarantee the accuracy, to begin with. So let’s see what can be done to look for pictures that closely resemble each other, but they can’t be called duplicates.