School of Computer Science, Carnegie Mellon University.
J Cogn Neurosci. 1990 Fall;2(4):320-43. doi: 10.1162/jocn.1990.2.4.320.
Significant progress has been made in understanding vision by combining computational and neuroscientific constraints. However, for the most part these integrative approaches have been limited to low-level visual processing. Recent advances in our understanding of high-level vision in the two separate disciplines warrant an attempt to relate and integrate these results to extend our understanding of vision through object representation and recognition. This paper is an attempt to contribute to this goal, by using a computational framework arising out of computer vision research to organize and interpret human and primate neurophysiology and neuropsychology.
通过结合计算和神经科学的约束条件，人们在理解视觉方面取得了重大进展。然而，在大多数情况下，这些综合方法仅限于低级视觉处理。在这两个独立学科中，我们对高级视觉的理解的最新进展值得尝试将这些结果联系和整合起来，通过对象表示和识别来扩展我们对视觉的理解。本文试图通过使用计算机视觉研究中出现的计算框架来组织和解释人类和灵长类动物的神经生理学和神经心理学，为实现这一目标做出贡献。
