Computing the Stereo Matching Cost with a Convolutional Neural Network Seminar Recent Trends in 3D Computer Vision
نویسندگان
چکیده
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 stereo matching, discusses the proposed method and shows results from recent stereo datasets.
منابع مشابه
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 کاملStereo 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 کاملPyramid Stereo Matching Network
Recent work has shown that depth estimation from a stereo pair of images can be formulated as a supervised learning task to be resolved with convolutional neural networks (CNNs). However, current architectures rely on patch-based Siamese networks, lacking the means to exploit context information for finding correspondence in illposed regions. To tackle this problem, we propose PSMNet, a pyramid...
متن کامل3D 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 کاملDeep Stereo Matching with Dense CRF Priors
Stereo reconstruction from rectified images has recently been revisited within the context of deep learning. Using a deep Convolutional Neural Network to obtain patchwise matching cost volumes has resulted in state of the art stereo reconstruction on classic datasets like Middlebury and Kitti. By introducing this cost into a classical stereo pipeline, the final results are improved dramatically...
متن کامل