Monday, April 14, 2008

How difficult is Vision?

Lately I have been wondering about the problem of Vision and how difficult it should be compared to problem of Artificial General Intelligence.
It seems to me that, given the order that it happened in Nature, processing visual input should be much simpler than using language or reasoning. I say this because there are quite simple animals with eyes, say a fish, a frog or a mouse... As I am not a biologist or neurologist, I am not sure what kind of visual tasks these animals are able to perform. For example, can a mouse tell if there is a cat in a picture or not?
In any case, I guess that these neuronal systems, much simpler than the human brain, are able to solve tasks that we have not yet achieved with Computer Vision algorithms.

If that's the case, I have two questions to my readers, who hopefully can help me clarify these issues:

- What is the "perfect" biological system to understand vision? It should be powerful enough to solve problems that we are interested in, such as distinguishing between different objects, but it should also have a relatively simple brain. Any ideas?

- If animals without human-level intelligence use vision quite effectively, does this mean that Artificial Intelligence will follow the same order of achievements? Or given the properties of computers, it will turn out to be easier to do reasoning, planning or even language processing?

Looking forward to reading your comments.

1 comment:

  1. Hello Hugo, nice blog.

    I see vision as a complex as needed data aquisition system.

    Based on the needed detail about your environment, evolution made you with a bettor or poor vision.

    If you need to identify all little details about your environment you should have a huge internal aggregated "catalogue" (memory). This catalogue is multidimensional and aggregate forms smells, feelings, etc etc over one notion and should have several levels in each dimension. (a cat has legs, legs have nails, etc). I see this catalogue as data warehouse house like system ( you can use SAP BW for that eheh)

    On top of this catalogue you have reasoning (don’t know scientific term) and the information should flow in both directions, you use the catalogue to decisions (intelligent or not) and based on the output you adapt this catalogue.

    “can a mouse tell if there is a cat in a picture or not”

    I think yes, it has enough good vision to identify a shape as a “cat” and it’s native reasoning tell him to do a lot of task, but this part is no vision anymore.

    “I guess that these neuronal systems, much simpler than the human brain, are able to solve tasks that we have not yet achieved with Computer Vision algorithms”

    I guess that too because animals vision “algorithm” should not be static, it evolutes, based on the evolution of that “catalogue” feeded by others data acquisition systems.

    Be cool dude 

    ReplyDelete