Abstract
We consider the representational capabilities of systems of receptive fields found in early mammalian vision, under the assumption that the successive stages of processing remap the retinal representation space in a manner that makes objectively similar stimuli (such as different views of the same 3D object) closer to each other, and dissimilar stimuli farther apart. We present theoretical analysis and computational experiments that compare the similarity between stimuli as they are represented at the successive levels of the processing hierarchy, from the retina to the nonlinear cortical units. Our results indicate that the representations at the higher levels of the hierarchy are indeed more useful for the classification of natural objects such as human faces.
| Original language | English |
|---|---|
| Pages (from-to) | 19-41 |
| Number of pages | 23 |
| Journal | Network: Computation in Neural Systems |
| Volume | 6 |
| Issue number | 1 |
| DOIs | |
| State | Published - 1995 |
| Externally published | Yes |
Bibliographical note
Funding Information:Part of this research was completed when YW was at Tel Aviv University. We thank Yael Moses for discussing with us some of her unpublished results, and for making available to us her database of face images, and E Adelson for comments on a previous version of this manuscript. The support of the Grodetzky Center for the Study of Higher Brain Function at the Weizmann Institute of Science is gratefully acknowledged. YW is supported by a training grant from the National Institute of General Medical Sciences. SE is an incumbent of the Sir Charles Clore Career Development Chair at the Weizmann Institute.