Joymo livecaster pro

Joymo’s Embedded system

Analysis that previously took days of manual work is now available to us automatically in seconds.

For the last three years, Joymo has been developing a remote broadcasting camera solution we call, internally, “V3” or “Version 3”. Now it is known as the Joymo LiveCaster Pro.

The Joymo LiveCaster Pro is an easy to install, 39-megapixel, 4K broadcasting camera. Sport teams and clubs across Norway (and the world) are invited to install the LiveCaster at their venues, set it up with your smart phone, and film or livestream all of their games and trainings in a safe, secure way. No need for camera people, no need extra work, fuss or bother. All delivered instantly and to the highest possible quality.

Arnstein Kleven is a senior member of Joymo’s development team, working on what is called an Embedded System. He’d like to explain what Joymo has done and what it means for innovation in Norway and for sports…

“Although we embedded developers still flock to blinking LEDs as moth flock to lights, embedded system development is so much more than that today. Wikipedia defines an embedded system as a computer system—a combination of a computer processor, computer memory, and input/output peripheral devices—that has a dedicated function within a larger mechanical or electrical system. Throw in a persistent network connection to this system and you can proudly label your system with one of the most prominent buzzwords of today: IoT (Internet of Things)”.

At Joymo we are at the forefront of IoT development and our system has multiple image sensors and 256 Nvidia CUDA cores, bringing computer vision and AI to the edge. This means we are able to do analysis and image processing at the location where it is recorded, and the network bandwidth can be utilized solely for delivering the best possible quality video streams to our cloud backend, which in turn serves them to our end-users.

Joymo work scaled

The developers in Joymo have a passion for technology, and we are using modern tools like the Rust programming language for writing our software, Yocto for creating our own custom Linux operating system, and CUDA for optimizing our advanced video pipeline.

With our IoT device and cloud backend as a base, the future possibilities are endless. With a growing number of games streamed to our cloud backend, we can use big data, AI and computer vision to learn more about the sports we all love. What are the keys to winning games? What positioning on the field will give you the most wins? Analysis that previously took days of manual work is now available to our customers automatically in seconds. 

Analysis that previously took hours and hours of manual work is now available to us automatically in seconds. With a highly scalable system, it is possible to combine several cameras to create a 3D mapping of the game in VR.