[top of page 9 of handout]It is often emphasised that sensory processes, especially vision, must allow animals to recognise real objects in the natural world and that this requires complex information processing (Bernays & Wcislo, 1996; Logothetis & Sheinberg, 1996; Riesenhuber and Poggio, 2000; Tarr and Cheng, 2003).
“Imagine waiting for incoming passengers at the arrival gate at the airport. Your visual system can easily find faces and identify whether one of them is your friend’s. As with other tasks that our brain does effortlessly, visual recognition has turned out to be difficult for computers. In its general form, it is a very difficult computational problem, which is likely to be significantly involved in eventually making intelligent machines. Not surprisingly, it is also an open and key problem for neuroscience.
The main computational difficulty is the problem of variability. A vision system needs to generalize across huge variations in the appearance of an object such as a face, due for instance to viewpoint, illumination or occlusions. At the same time, the system needs to maintain specificity. It is important here to note that an object can be recognized at a variety of levels of specificity: a cat can be recognized as “my cat” on the individual level, or more broadly on the categorical level as “cat”, “mammal”, “animal” and so forth. (Riesenhuber and Poggio, 2000; p. 1199)
"The ability to readily adapt to novel situations requires something beyond storing specific stimulus-response associations. Instead, many animals can detect basic characteristics of events and store them as generalized classes. Because these representations are abstracted beyond specific details of sensory inputs and motor outputs, they can be easily generalized and adapted to new circumstances. Explorations of neural mechanisms of sensory processing and motor output have progressed to the point where studies can begin to address the neural basis of abstract, categorical representations. Recent studies have revealed their neural correlates in various cortical areas of the non-human primate brain." Abstract of Miller et al., (2003)
“Visual recognition of objects is an impressively difficult problem that biological systems solve effortlessly.” (Tarr and Cheng, 2003)
“Visual object recognition is of fundamental importance to most animals.” (Logothetis and Sheinberg, 1996)
Categorization and pattern recognition, involving grouping of sets of stimuli in functional relations to responses, appear to be consequences of perceptual learning. (Overall conclusion, p. 4 of handout)
[bottom of page 9 of handout]Comparative anatomy and physiology of the visual system (e.g. Walker, 1985 pp 266-286)