The state of Computer Vision and AI: we are far away

Computer vision and AI face enormous challenges when tasked with understanding images the way humans naturally do, as demonstrated by a photo of Obama secretly adding weight to someone's scale. While humans instantly process complex layers of information—recognizing people through mirrors, understanding 3D physics, predicting emotional reactions, and reasoning about social dynamics—current AI systems can only handle basic tasks like image classification. The gap between today's computer vision capabilities and human-level scene understanding reveals we lack fundamental knowledge about gathering appropriate training data and developing algorithms capable of such sophisticated reasoning.