Widening siamese architectures for stereo matching

11/01/2017
by   Patrick Brandao, et al.
0

Computational stereo is one of the classical problems in computer vision. Numerous algorithms and solutions have been reported in recent years focusing on developing methods for computing similarity, aggregating it to obtain spatial support and finally optimizing an energy function to find the final disparity. In this paper, we focus on the feature extraction component of stereo matching architecture and we show standard CNNs operation can be used to improve the quality of the features used to find point correspondences. Furthermore, we propose a simple space aggregation that hugely simplifies the correlation learning problem. Our results on benchmark data are compelling and show promising potential even without refining the solution.

READ FULL TEXT

page 2

page 4

page 5

page 6

research
10/28/2022

Comparison of Stereo Matching Algorithms for the Development of Disparity Map

Stereo Matching is one of the classical problems in computer vision for ...
research
05/13/2019

FPGA-based Binocular Image Feature Extraction and Matching System

Image feature extraction and matching is a fundamental but computation i...
research
01/28/2019

Fast Hierarchical Depth Map Computation from Stereo

Disparity by Block Matching stereo is usually used in applications with ...
research
09/17/2022

Active-Passive SimStereo – Benchmarking the Cross-Generalization Capabilities of Deep Learning-based Stereo Methods

In stereo vision, self-similar or bland regions can make it difficult to...
research
09/10/2020

Long Range Stereo Matching by Learning Depth and Disparity

Stereo matching generally involves computation of pixel correspondences ...
research
01/12/2018

Deep Stereo Matching with Explicit Cost Aggregation Sub-Architecture

Deep neural networks have shown excellent performance for stereo matchin...
research
01/13/2020

Deep Learning Stereo Vision at the edge

We present an overview of the methodology used to build a new stereo vis...

Please sign up or login with your details

Forgot password? Click here to reset