PhotonVision is a free, fast, and easy-to-use vision processing solution for the FIRSTRobotics Competition. PhotonVision is designed to get vision working on your robot quickly, without the significant cost of other similar solutions. Using PhotonVision, teams can go from setting up a camera and coprocessor to detecting and tracking targets by simply tuning sliders. With an easy to use interface, comprehensive documentation, and a feature rich vendor dependency, no experience is necessary to use PhotonVision. No matter your resources, using PhotonVision is easy compared to its alternatives.
PhotonVision has a myriad of advantages over similar solutions, including:
Compared to alternatives, PhotonVision is much cheaper to use (at the cost of your coprocessor and camera) compared to alternatives that cost $400. This allows your team to save money while still being competitive.
Easy to Use User Interface
The PhotonVision user interface is simple and modular, making things easier for the user. With a simpler interface, you can focus on what matters most, tracking targets, rather than how to use our UI. A major unique quality is that the PhotonVision UI includes an offline copy of our documentation for your ease of access at competitions.
PhotonLib Vendor Dependency
The PhotonLib vendor dependency allows you to easily get necessary target data (without having to work directly with NetworkTables) while also providing utility methods to get distance and position on the field. This helps your team focus less on getting data and more on using it to do cool things. This also has the benefit of having a structure that ensures all data is from the same timestamp, which is helpful for latency compensation.
Using PhotonVision allows the user to calibrate for their specific camera, which will get you the best tracking results. This is extremely important as every camera (even if it is the same model) will have it’s own quirks and user calibration allows for those to be accounted for.
High FPS Processing
Compared to alternative solutions, PhotonVision boasts higher frames per second which allows for a smoother video stream and detection of targets to ensure you aren’t losing out on any performance.
PhotonVision provides low latency processing to make sure you get vision measurements as fast as possible, which makes complex vision tasks easier. We guarantee that all measurements are sent from the same timestamp, making life easier for your programmers.
Fully Open Source and Active Developer Community
You can find all of our code on GitHub, including code for our main program, documentation, vendor dependency (PhotonLib), and more. This helps you see everything working behind the scenes and increases transparency. This also allows users to make pull requests for features that they want to add in to PhotonVision that will be reviewed by the development team. PhotonVision is licensed under the GNU General Public License (GPLv3) which you can learn more about here.
You can use multiple cameras within PhotonVision, allowing you to see multiple angles without the need to buy multiple coprocessors. This makes vision processing more affordable and simpler for your team.
Using our comprehensive documentation, you will be able to easily start vision processing by following a series of simple steps.