The Khronos OpenVX API offers a set of optimized primitives for low-level image processing, computer vision and neural network operators. It provides a simple method for writing optimized code that is portable across multiple hardware vendors and processors, including CPUs, GPUs and special-function hardware. In this talk, we will introduce OpenVX 2.0, the latest version of the standard, and explain its key improvements. Using real-world use cases from Bosch, Texas Instruments and others, we’ll show how OpenVX is used to build, verify and coordinate computer vision and neural network graph executions, enabling software developers to spend more time on algorithmic innovations without worrying about the performance and portability of their applications.