Quick Start

This section provides a brief description of how to quickly start running this project. For more detailed environment configuration and compilation methods, see Build and Run.

Environment Dependencies

Before you start, make sure you have installed the following tools:

• C++ compiler (supports C++11 or higher)

• CMake

• OpenCV

Data Preparation

This project supports the KITTI dataset by default. Please place the left and right camera images in a folder with the following structure:

data/
├── left
│ ├── 0000000000.png
│ ├── 0000000001.png
│ └── ...
└── right
├── 0000000000.png
├── 0000000001.png
└── ...

If you need to use other datasets, please make sure that the file naming and path method conform to the code logic.

Quick Run

1. Clone the code and enter the project directory:

   git clone
   cd Modelforge

2. Compile and run:

   mkdir build && cd build
   cmake ..
   make -j4
   ./Modelforge

   After the program runs, it will output the frame number and open a visualization window (such as disp8, result, stixel) to display the disparity map and columnar pixel results.

Common Problems

• If it prompts that OpenCV is not found, please check the OpenCV installation path or CMake configuration.

• Incorrect dataset path or file naming will also cause image reading failure.