Lacmus Application

Cross-platform application that searches for lost people in the forest using object detection neural networks. The application runs on Linux, Windows and Mac OS X. Lacmus application written in C#, .NET Core, ReactiveUI and Avalonia and use Tensorflow / ONNX for inference deep learning Models.

Main view

Docs

You can find more documentation here: Rus Eng

System requirements and supported platforms

Windows 7 SP1 or higher (64 bit) - supports CPU and GPU (Cuda, DirectML) ml inference.
Linux (64 bit) - supports CPU, GPU (Cuda, ROCm) and Coral Edge TPU ml inference.
MacOS 11 or higher (64 bit) - supports CPU ml inference.
Minimum 2x CPU cores and 4 GB of RAM for ML inference.

Brief description of architecture

Architecture diagram

Application use LacmusPlugins for ML model inference and PluginReposiroty web API for hosting ML models.

Building from source

$ git clone https://github.com/lacmus-foundation/lacmus-app.git
$ cd lacmus-app/src
$ dotnet restore
$ dotnet run --project LacmusApp.Avalonia

Contributing

Contributors are welcome! You can contact me (gosha20777@live.ru) to get more info.