Flexible embedded vision processing architectures for machine-learning applications
Machine-learning strategies for embedded vision are evolving so quickly that designers need access to flexible, heterogenous processor architectures that can adapt as the algorithms evolve.
