VisionPortal Overview

FIRST Tech Challenge introduces VisionPortal, a comprehensive new interface for vision processing.

  • For FTC Blocks and Java teams, VisionPortal offers key capabilities of AprilTag and EasyOpenCV, along with TensorFlow Object Detection (TFOD) – at the same time!

    Dual Detection

    Dual Preview with both AprilTags and TensorFlow


  • AprilTag detections include ID code and pose: tag location and orientation, relative to the camera.

  • Camera Controls, which can improve AprilTag and TFOD performance for webcam, are now fully available to FTC Blocks users.

  • Multiple cameras can operate at the same time – phone camera and/or webcam.

    Dual Camera

    Multiple Camera View


  • Sample OpModes and new tools are available to operate and customize these features, including the Builder pattern.

  • For heavy video processing, many options are available to manage CPU resources and USB bandwidth.

  • DS and RC previews can be BIG!

    Full Screen

    Full Screen Preview

Many other new and improved features await your discovery in VisionPortal and beyond.


In preparation for the 2023-2024 CENTERSTAGE season, the new Software Development Kit (SDK) VisionPortal includes built-in support for AprilTag technology. Previously, Teams needed to download and incorporate external libraries, complicating the programming effort.

AprilTag is a popular vision technology for detecting a simple black-and-white tag, used to estimate position and orientation. In the 2022-2023 POWERPLAY game, many Teams enjoyed AprilTag’s reliable Autonomous performance for Signal Sleeve recognition.

Dual Detection

Photo Credit: Mike Silversides

All sections of this Guide assume prior reading of the AprilTag Introduction .

The SDK describes AprilTag pose relative to the camera, by default. This computing process is called pose estimation, a term that emphasizes this is an estimate only, based on many factors including camera calibration. You must determine AprilTag’s best use for reaching your goals.


Much credit to

Questions, comments and corrections to westsiderobotics@verizon.net