Throwing Down the Visual Intelligence Gauntlet

In recent years, scientific and technological advances have produced artificial systems that have matched or surpassed human capabilities in narrow domains such as face detection and optical character recognition. However, the problem of producing truly intelligent machines still remains far from being solved. In this chapter, we first describe some of these recent advances, and then review one approach to moving beyond these limited successes – the neuromorphic approach of studying and reverse-engineering the networks of neurons in the human brain (specifically, the visual system). Finally, we discuss several possible future directions in the quest for visual intelligence. 1 Artificial Intelligence: Are We There Yet? Every few years (and sometimes more often), the world becomes abuzz with excitement over some new technology that, finally, after all these years, promises to fulfill the dream of Artificial Intelligence (AI), first proposed back in 1956 at the dawn of the Age of Computing. The latest of these technologies is Watson, a natural language question-answering computer system that, in February 2011, defeated two of the best human contestants ever in the quiz show Jeopardy! In 2007, the buzz-worthy news was that six teams successfully completed the DARPA Grand Challenge [3], a simulated 60-mile urban course designed to test the capabilities of driverless vehicle systems. The overall winner completed the course just over 4 hours, averaging approximately 14 mph. Just two years prior, in 2005, five teams completed the 132-mile off-road desert course, a stunning reversal of the 2004 results, in which the best vehicle only managed to complete an embarrassing 7.36 miles of the 150-mile course [3]. Apart from these and other headline-grabbing milestones like chess-playing computer Deep Blue defeating the reigning world chess champion in 1997, there have also been a host of less-publicized technological advances that have matched (and sometimes even surpassed) human abilities. These include face-detection technology in digital cameras [4], a pedestrian-detection feature in the latest luxury car models [1, 6], and optical character recognition (OCR) technology used by postal services [9] around the world, just to name a few.