A short lecture-style video section explaining why computer vision is hard: humans find object recognition (e.g., identifying a shirt) effortless, while computers must infer object identity from raw pixel values. The speaker credits machine learning and deep learning for recent improvements but emphasizes the unintuitive gap between human and machine perception.
One of the non intuitive things about vision is that it's so easy for a person to look at you and say, oh, you're wearing a shirt. It's so hard for a computer to figure it out. And because it's so easy for humans to recognize objects, it's almost difficult to understand why this is a complicated thing for a computer to do. You know, what the computer has to do is look at all, look at all numbers, all the pixel brightness values, say, look at all these numbers, say, oh, these numbers correspond to, to a black shirt. And it's amazing that with machine learning, deep learning, computers are getting really good at this, right?
