Mind Matters Natural and Artificial Intelligence News and Analysis

TagFrankensteins in machine vision testing

Busy evening cityscape with cars and people on 42nd Street in Midtown Manhattan New York City

One Way Human Vision Is Better Than a Machine’s

The problem machine vision has with understanding what things *should* look like creates risks for traffic video safety systems, researchers say

Because machine vision absorbs information without context, it simply doesn’t “see” what a human sees — and the results could be “dangerous in real-world AI applications,” warns York University prof James Elder. He and his colleagues did an interesting experiment with deep convolutional neural networks (DCNNs). They used “Frankensteins” — models of life forms that are distorted in some way — to determine how they would be interpreted by humans or by machine vision: “Frankensteins are simply objects that have been taken apart and put back together the wrong way around,” says Elder. “As a result, they have all the right local features, but in the wrong places.” The investigators found that while the human visual system is confused by Read More ›