A model of motion, stereo, and monocular depth perception

0Citations
Citations of this article
3Readers
Mendeley users who have this article in their library.
Get full text

Abstract

Visual cortical processing is segregated into pathways each consisting of several cortical areas. We identified key mechanisms of local competitive interaction, feedforward integration and modulatory feedback as common principles of integration and segregation of ambiguous information to implement a principle of evidence accumulation and feedback hypothesis testing and correction. In a previous work we demonstrated that a model of recurrent VI-MT interaction disambiguates motion estimates by filling-in. Here we show that identical mechanisms along the ventral VI-V2-V4 pathway are utilized for the interpretation of (1) stereoscopic disparity and (2) relative depth segregation of partially overlapping form. The results show that absolute and relative depth ambiguities are resolved by propagation of sparse depth cues. Lateral inhibition emerges at locations of unambiguous information and initiates the recurrent disambiguation process. Our simulations substantiate the proposed model with key mechanisms of integration and disambiguation in cortical form and motion processing. © Springer-Verlag 2004.

Cite

CITATION STYLE

APA

Bayerl, P., & Neumann, H. (2004). A model of motion, stereo, and monocular depth perception. Lecture Notes in Computer Science (Including Subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), 3175, 95–102. https://doi.org/10.1007/978-3-540-28649-3_12

Register to see more suggestions

Mendeley helps you to discover research relevant for your work.

Already have an account?

Save time finding and organizing research with Mendeley

Sign up for free