Paul Kruszewski, President of WRNCH, presents the "Democratizing Computer Vision Development: Lessons from the Video Game Industry" tutorial at the May 2016 Embedded Vision Summit.
Computer vision offers great promise: algorithms are maturing rapidly and processing power continues to grow by leaps and bounds. But today’s approach to computer vision software development – hiring a team of computer vision PhDs to hack OpenCV – is not scalable and represents a major bottleneck to mass deployment of vision-enabled products. The video game industry faced similar challenges in the early 2000s, when it became impractical for game developers to write an entire game engine from scratch.
Today, small teams of independent game developers leverage commercial game engines like Unity to build complex video games that only five years ago would have required a 100+ person team. In this talk, based on experience in the game industry, Kruszewski predicts how computer vision software development will evolve to enable creation of thousands of new vision-enabled products. He also shares ways that managers and entrepreneurs can avoid the most serious pitfalls of vision software development today.