Movidius

Movidius was a fabless semiconductor company specializing in the design of ultra-low-power vision processing units (VPUs) for edge computing and embedded artificial intelligence applications. Co-founded in 2005 by Sean Mitchell, David Moloney, and Val Muresan in Dublin, Ireland, the company was headquartered in San Mateo, California, and was acquired by Intel in September 2016. Wikipedia Between 2006 and 2016, Movidius raised nearly $90 million in capital funding. Wikipedia The company's core products are its Myriad family of VPUs. The Myriad 2 chip is a manycore vision processing unit designed to function on power-constrained devices, capable of running at between 80 and 150 GFLOPS on just 1 watt of power. Wikipedia The third-generation Myriad X, built on 16 nm process technology, integrates a dedicated Neural Compute Engine and can process up to 4 trillion operations per second with a thermal design power of only 1.5 watts. Assured Systems Movidius also developed the Intel Neural Compute Stick, a USB-based development tool that enables prototyping of deep learning applications without cloud connectivity, supporting frameworks such as TensorFlow and Caffe. Wikipedia Movidius serves OEMs and developers building AI-powered products across drones, robotics, security cameras, AR/VR headsets, industrial machine vision, and IoT devices. Notable integrations include Google's Project Tango, the Google Clips camera, DJI Mavic consumer drones, and the Intel RealSense Tracking Camera T265. Wikipedia The company's value proposition centers on enabling real-time, on-device deep learning inference at the network edge, eliminating the latency, bandwidth, and energy costs of cloud-based processing. Its chips are well suited for applications in surveillance, retail, and transportation where power efficiency is critical. Macnica Americas Intel continues to sell products under the Movidius brand and has integrated the VPU technology into its broader OpenVINO toolkit ecosystem. Keywords: computer vision, vision processing unit, VPU, edge AI, deep learning, neural compute engine, Myriad 2, Myriad X, embedded AI, low-power processors, fabless semiconductor, Intel acquisition, drones, IoT, machine vision, neural compute stick, inference at the edge, TensorFlow, OpenVINO, AR/VR, robotics, smart cameras