Neural architectures for stereo vision

نویسندگان

  • Andrew J Parker
  • Jackson E T Smith
  • Kristine Krug
چکیده

Stereoscopic vision delivers a sense of depth based on binocular information but additionally acts as a mechanism for achieving correspondence between patterns arriving at the left and right eyes. We analyse quantitatively the cortical architecture for stereoscopic vision in two areas of macaque visual cortex. For primary visual cortex V1, the result is consistent with a module that is isotropic in cortical space with a diameter of at least 3 mm in surface extent. This implies that the module for stereo is larger than the repeat distance between ocular dominance columns in V1. By contrast, in the extrastriate cortical area V5/MT, which has a specialized architecture for stereo depth, the module for representation of stereo is about 1 mm in surface extent, so the representation of stereo in V5/MT is more compressed than V1 in terms of neural wiring of the neocortex. The surface extent estimated for stereo in V5/MT is consistent with measurements of its specialized domains for binocular disparity. Within V1, we suggest that long-range horizontal, anatomical connections form functional modules that serve both binocular and monocular pattern recognition: this common function may explain the distortion and disruption of monocular pattern vision observed in amblyopia.This article is part of the themed issue 'Vision in our three-dimensional world'.

برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

منابع مشابه

Segmentation and Location Computation of Bin Objects

In this paper we present a stereo vision based system for segmentation and location computation of partially occluded objects in bin picking environments. Algorithms to segment partially occluded objects and to find the object location [midpoint,x, y and z co‐ordinates] with respect to the bin area are proposed. The z co‐ ordinate is computed using stereo images and neural ...

متن کامل

Cascaded multi-scale and multi-dimension convolutional neural network for stereo matching

Convolutional neural networks(CNN) have been shown to perform better than the conventional stereo algorithms for stereo estimation. Numerous efforts focus on the pixel-wise matching cost computation, which is the important building block for many start-of-the-art algorithms. However, those architectures are limited to small and single scale receptive fields and use traditional methods for cost ...

متن کامل

Stereo Matching by Training a Convolutional Neural Network to Compare Image Patches

We present a method for extracting depth information from a rectified image pair. Our approach focuses on the first stage of many stereo algorithms: the matching cost computation. We approach the problem by learning a similarity measure on small image patches using a convolutional neural network. Training is carried out in a supervised manner by constructing a binary classification data set wit...

متن کامل

Computing the Stereo Matching Cost with a Convolutional Neural Network Seminar Recent Trends in 3D Computer Vision

This paper presents a novel approach to the problem of computing the matching-cost for stereo vision. The approach is based upon a Convolutional Neural Network that is used to compute the similarity of input patches from stereo image pairs. In combination with state-ofthe-art stereo pipeline steps, the method achieves top results in major stereo benchmarks. The paper introduces the problem of s...

متن کامل

3D Reconstruction Approach based on Wavelet Analysis and Neural Network in Neuro-vision System

In this paper, a 3D reconstruction approach based on wavelet analysis and neural networks is presented. It can be used in neuro-vision system. The approach can be divided into two parts. First, the stereo matching problem is solved with wavelet analysis. Dyadic discrete wavelet analysis is adopted in this process and stereo matching process is realized with global optimization. A coherent hiera...

متن کامل

ذخیره در منابع من


  با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید

برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

عنوان ژورنال:

دوره 371  شماره 

صفحات  -

تاریخ انتشار 2016