There’s a new AI robot called ‘There’s Waldo’ that has been built and programmed to find Waldo and literally point him out. The robot once initialized is instructed to extend and take a photo of the canvas (‘Where’s Waldo book). It then uses OpenCV to source and extract faces from the sample photo. The faces are sent to the Google Auto ML Vision service which compares each one against the trained Waldo model. If ‘There’s Waldo” confidently finds a match of at least 95% (0.95) or more then it will extend its arm to the coordinates of the matching face and point it out. If there are multiple Waldos in a photo it will point to each one. Watch ‘There’s Waldo’ in action below.