This project provides the UFEEL native library, a tool to handle emotion recognition, eye tracking, speech to text and heart rate sensing.
git clone https://github.com/UFeel-dev/ufeel
cd ufeel
git lfs install
git submodule update --init --recursive
git lfs pull
cd docker
docker build -t ufeeldocker .
cd ..
./docker/run.sh
./build.sh all
Docker is the recommended build environment. It provides a reproducible toolchain and avoids requiring the native build dependencies on the host.
Note: you will need appropriate input devices to test some features, such as a camera and a microphone.
To use clang-related tools (e.g. linting, formatting, or IDE features), set up the compilation database:
bear.build directory if it exists.bear -- ./build.sh all
The generated compilation database from CMake is not always directly usable by all tooling.
Manual builds are supported but require compatible versions of all third-party dependencies.
The native package still relies on some operating-system libraries, including glibc, libstdc++, audio libraries and other system dependencies.
If some dependencies are missing for your system, please open an issue. The Dockerfile documents the build environment.
This project uses Git LFS to store large third-party libraries in 3rdparty/, including prebuilt .so and .a files.
yaml-cpp is maintained as a Git submodule.
After cloning, initialize all submodules and download the Git LFS files:
git lfs install
git submodule update --init --recursive
git lfs pull
.
├── build.sh
├── csharp/ # C# test project
├── Library/ # Native sources
├── Wrapper/ # C++ wrapper
├── 3rdparty/ # External dependencies
├── build/ # Generated build files
└── docker/ # Docker environment
If you prefer running the steps manually:
./build.sh configure
./build.sh build
./build.sh package
| Document | Description |
|---|---|
| Contributing | Contribution guidelines |
| Disclaimer | Limitations and prohibited use |
| Docker | Docker setup instructions |
Apache 2.0 License - see LICENSE.