RobA1701 This is exactly how it should work - iterative process. Program gives you facial suggestions in small batches and you confrim and reject. Program learns from your input to provide more suggestions. At first you will mostly confirm, but not for long. Once program will exhaust all obvious mathces, it will start providing you with more challegning suggestions, and here is where you have to be very careful. Incorrect confirmations will end up in stream of incorrect suggestions. And the more mistakes you make the lower is suggestion quality. That is why it is done in small batches - to make it manageable, and learn from rejections earlier, so program will not overwhelm you with lots of incorrect suggestions early on in the process.
