Wednesday, February 12, 2020

Bridging the gap between human and machine vision

Recommendable! Despite tremendous success and progress over the past 8 years, computer vision is still far inferior to human vision.

"In particular, it is possible that the invariance with one-shot learning exhibited by biological vision requires a rather different computational strategy than that of deep networks. ... "A key reason for this difference is the relative invariance of the primate visual system to scale, shift, and other transformations. Strangely, this has been mostly neglected in the AI community, in part because the psychophysical data were so far less than clear-cut. Han's work has now established solid measurements of basic invariances of human vision.”"

Bridging the gap between human and machine vision | MIT News: MIT researchers led by Yena Hang have developed a more robust machine-vision architecture by studying how human vision responds to changing viewpoints of objects.

Here is the underlying research paper:
Scale and translation-invariance for novel objects in human vision (Nature, Open Access)

No comments: