DEV Community: aposb

Basic Stereo Algorithms (Evolution)

aposb — Fri, 05 Jun 2026 13:32:19 +0000

Stereo matching is a fundamental problem in computer vision and understanding how different approaches relate to one another can provide valuable insight into the field. In this post, I’m sharing a repository that presents simplified implementations of several classic stereo matching algorithms (in MATLAB and Python), making it easier to study their underlying concepts and compare their implementations.

The project includes Block Matching, Dynamic Programming, Semi-Global Matching and two variants of Belief Propagation (Directional and Synchronous). Rather than focusing solely on the final algorithms, the implementations have been simplified and adapted to highlight the connections between them. This allows readers to follow how ideas from one method can lead to the next and gain a clearer understanding of the evolution of stereo matching techniques.

Repository: Basic Stereo Algorithms (Evolution)

Stereo Matching Algorithms in MATLAB and Python

aposb — Tue, 03 Feb 2026 13:33:34 +0000

Stereo matching is a core problem in computer vision, and performance matters, especially when working with large images or real-time systems. In this post, I’m sharing a set of fast, optimized stereo matching algorithms implemented in MATLAB and Python.

What's Included

Two versions of Block Matching
Two versions of Dynamic Programming
Semi-Global Matching and Semi-Global Block Matching
Three versions of Belief Propagation

All algorithms are available in: Stereo Matching Algorithms in MATLAB and Python

Example Outputs

The algorithms are tested using the Tsukuba stereo image.

Here are the resulting disparity maps generated by each method.

Block Matching using Sum of Absolute Differences

Block Matching using Census Transformation

Dynamic Programming with Left-Right Axes DSI

Dynamic Programming with Left-Disparity Axes DSI

Semi-Global Matching

Semi-Global Block Matching

Belief Propagation with Accelerated Message Update Schedule

Belief Propagation with Synchronous Message Update Schedule

There is also a second approach to Belief Propagation with Synchronous Message Update Schedule, which produces the same result.

Integrating OpenCV (C++) with Visual Studio 2019 - the proper way

aposb — Wed, 02 Jul 2025 09:12:01 +0000

In this post, I will set up OpenCV v4.10.0 on Windows 10 and create a demo C++ project to demonstrate how to integrate OpenCV with Visual Studio 2019.

Setup OpenCV

1. Download

Go to the OpenCV releases page (https://opencv.org/releases/).
Find version 4.10.0.
Click Windows to download the installer.

2. Install

Run the installer.
Specify "C:\" as the installation path.
After installation, a new folder named opencv should appear in "C:\".

3. Set Environment Variables

Open Advanced System Properties in Windows 10.
Click Environment Variables...
Create a new system variable named OPENCV_DIR with the value: C:\opencv\build\x64\vc16

In the system variable Path, add: %OPENCV_DIR%\bin

Create the Demo Project

1. New Project

Create a new project in Visual Studio.
Set the programming language to C++.
From the filtered list, select Console App.

2. Project Properties

Open the Property Pages window (Project > Properties).

Debug configuration:

Set Configuration to Debug and Platform to x64.
In C/C++ > General, under Additional Include Directories, add: $(OPENCV_DIR)\..\..\include

In Linker > General, under Additional Library Directories, add: $(OPENCV_DIR)\lib

In Linker > Input, under Additional Dependencies, add: opencv_world4100d.lib;

Click Apply.

Release configuration:

Set Configuration to Release.
Repeat the same steps for include and library directories.
In Linker > Input, under Additional Dependencies, add: opencv_world4100.lib;
Click Apply.

Finally, in the Configuration Manager, remove the x86 (Win32) platform.

Run the Program

Replace the default project code with the sample code below.

#include <opencv2/opencv.hpp>

using namespace cv;

int main()
{
    Mat image = Mat::zeros(300, 600, CV_8UC3);
    circle(image, Point(250, 150), 100, Scalar(0, 255, 128), -100);
    circle(image, Point(350, 150), 100, Scalar(255, 255, 255), -100);
    imshow("Display Window", image);
    waitKey(0);
    return 0;
}

Build and run the project. If everything is set up correctly, a window should appear.