Skip to content

travislhopkins/INVSC-janice

 
 

Repository files navigation

JanICE API

Overview

The JanICE API is a C API that provides a common interface between computer vision algorithms and agencies and entities that would like to use them.

Building

The API is header only by design and therefore cannot be "built". However, a build protocol designed around the CMake program is provided as a default way to build implementations. Additionally, a small test harness and associated command line utilties are included that provide examples to users of best practices for using the API.

The protocol is quite simple. 2 variables give the paths to the implementation and I/O implementation libraries. The implementation library is specified by JANICE_IMPLEMENTATION and the I/O library by JANICE_IO_IMPLEMENTATION. Each variable should give the full path to the respective library and can be specified to cmake using the -D option. For example, an implementation that has all functionality in a library fake_janice_implementation.so and all I/O functionality in a library fake_janice_io_implementation.so could be linked against using the following command:

$cmake -DJANICE_IMPLEMENTATION=/path/to/fake_janice_implementation.so \
       -DJANICE_IO_IMPLEMENTATION=/path/to/fake_janice_io_implementation.so ..

The .. at the end gives a relative path to the root of this repository. The test harness and command line executables automatically check if JANICE_IMPLEMENTATION is defined and will build if it is not an empty string.

License

This project is licensed under the MIT license and is free for academic and commercial use.

Documentation

The documentation for the API is hosted here. The complete documentation for the API is available in "docs" folder. A PDF file is provided for convienience. Alternatively, you can build the docs yourself with the Sphinx documentation generator.

Contributing

Contributions are generally welcome. Please open a pull request.

About

An open API for computer vision algorithms

Resources

License

Stars

Watchers

Forks

Packages

No packages published

Languages

  • C++ 87.9%
  • C 9.9%
  • CMake 2.2%