cpp-cv-project-template: A Python repository from changh95

"It's never been easier than this!"

Purpose 🔥

This repo contains a C++ project template for developing computer vision applications.

Features 🌟

Currently, this repo is going under a rigorous change! Stay tuned for a big update!

The project comprise a widely-used C++ project structure.
The project supports installing essential toolchains for C++ programming and debugging.
- git, build-essentials, cppcheck, cmake, clang, gcc, clang-tools, clang-tidy, lldb, lld, libc++, libomp (:heavy_check_mark:)
The project supports the following 3rdParty libraries:
- Python3 + basic packages (numpy, pandas, matplotlib, jupyter notebook, voila, tqdm, nbconvert) (:heavy_check_mark:)
- Open3D - Python only for now (:heavy_check_mark:)
- Eigen (:heavy_check_mark:)
- OpenCV (:heavy_check_mark:)
  - Non-free algorithms currently disabled
- Ceres-solver (:heavy_check_mark:)
- GTSAM (:heavy_check_mark:)
- Pangolin (:heavy_check_mark:)
- PCL (:heavy_check_mark:)
  - Visualization disabled
- OpenGL (:heavy_check_mark:)
The project supports various tools to make good C++ project practices.
- GTest (:heavy_check_mark:)
- spdlog (:heavy_check_mark:)
- fast-cpp-csv-parser (:heavy_multiplication_x:)
- nlohmann-json (:heavy_multiplication_x:)

Status legend: :heavy_check_mark: - Fully supported, :white_check_mark: - Partially supported / Build-only, :heavy_multiplication_x: - Not supported yet

How to use 📖

You need Python3 to use the automation scripts for project setup and build.

Linux

# Install Python 3 and required modules
sudo apt install python3 python3-pip
pip3 install pyyaml gitpython

# Edit `setup_config.yaml` to configure project
gedit ./thirdparty/packages.yaml

# Install dependencies
./build_dependencies.py

Edit the packages.yaml file to configure your project dependencies.
You can also use the optional --password YOUR_PASSWORD argument to avoid manually typing your Linux password for every internal sudo command usage.
You can also use the optional --d argument to also build debug libraries.
You can also use the optional --system argument to build inside the system.