Google Summer Of Code Java version setup.


The object detection Java library is based on a OpenCV DepthAI C++ library. It can run on Linux, Windows and OSX platforms. The hardware supports X86_64 and ARM based embedded devices. I used the FRC WPILibrary simulator to test the results of the person detector so that the RoboRIO hardware is not necessary. The detection results will be passed to the simulator or actual RoboRIO by using the FRC WPILibrary NetworkTables components.


  • FRC RoboRIO simulator
  • Ubuntu 20.04 on X86_64
  • Oracle Java 8 SDK
  • Maven 3.6.3
  • DepthAPI driver
  • OAK-D camera with USB connected to PC

Linux Installation

There are some required dependencies that are needed to be compiled and installed in order to simulate the results on a RoboRIO.

  • RoboRIO Simulator: Refer to this link for setup instructions. After a successful installation, the simulator should resemble the following:

  • JavaCpp for DepthAI: Refer to this link for build steps.
      cd javacpp-presets
      mvn install --projects .,depthai
      cd javacpp-presets/depthAI/platform
      mvn clean install

    After the above steps are complete, a depthai-platform.jar will be created and installed in the maven repository.

  • WPILibrary NetworkTable component
    • Since the version that I used has been removed from the maven repository, manual installation is necessary. To begin, download the NetworkTables library from this link.

    • Start a terminal and cd to the path where the jar file was saved. Once complete, run the below command to install this jar to the maven repository.

      mvn install:install-file -Dfile=./NetworkTables-3.1.7-20170802171912-5-gf43675e-desktop.jar -DartifactId=NetworkTables -Dversion=3.1.7 -Dpackaging=jar
  • java-ironoak install and run

      git clone
      mvn clean install
      mvn compile exec:java

    Make sure that the Oak-D is connected with the USB 3 port using the original cable. From the RoboRIO simulator, the person detector’s results have been passed to the NetworkTables window, and should display the values for when a person has been detected.

Java FRC demo