Myriad 2 delivers 20x gain in energy efficiency for computational imaging and visual awareness applications
San Mateo, California – July 30, 2014 – Movidius, the leader in mobile vision processor technology for connected devices, today announced Myriad 2, the next generation of its vision processor unit (VPU) system-on-a-chip (SoC). Compared to Myriad 1—which was introduced as part of Google’s Project Tango—the new chipset delivers 20x more processing efficiency in terms of computations per watt of power consumed. Myriad 2 comprises a larger and upgraded set of programmable processors as well as a new set of dedicated and configurable image and vision accelerators to power the next wave of computational cameras.
Myriad 2 is a programmable, high-performance, ultra-low power SoC that brings computational imaging and visual awareness to life for a variety of mobile and portable applications. Myriad 2 features are key to enabling use cases such as vision-enhanced image and video capture, 3D scanning and mapping, 360-degree panoramic video, indoor location-based services, immersive gaming and many new possibilities.
“As rapid innovations in mobile, wearable, and embedded vision applications continue, users expect increasingly sophisticated and immersive experiences that do not compromise the device’s battery life,” said Remi El-Ouazzane, CEO of Movidius. “The next disruption in mobile will be about incorporating vision applications into connected devices, bringing them closer to the complexity and sophistication of human vision. Thanks to Myriad 2’s radically innovative architecture, our customers can now create proprietary ultra-low power, captivating vision experiences that drive demand for existing products and accelerate completely new product categories.”
Key features of the Myriad 2 chipset include:
- Superior performance at ultra-low power: Myriad 2, implemented in 28-nanometer technology, is capable of performance exceeding two trillion 16-bit operations per second while consuming an average of less than 500 milliwatts. It supports up to six full HD 60 frames per second camera inputs simultaneously via 12 MIPI lanes.
- Programmability: Myriad 2 is equipped with 12 programmable vision-specific vector processors to enable rapid innovation and to enable device manufacturers to differentiate using their proprietary software.
- Low latency processing: Given its highly parallelized data processing architecture and its new and highly efficient on-chip memory fabric, Myriad 2 can achieve high-performance processing with notably low latency. As an example, for the standard vision processing benchmark known as the Haar Cascade classification, Myriad 2 can calculate 50,000 multi-scale classifications per high-definition (HD) video frame in a mere seven milliseconds.
- Flexible usage for different applications: Myriad 2 operates either as a co-processor to the main application processor in a vision-enabled mobile device, or as a standalone processor in wearable or embedded applications.
As software tools are the critical bridge between the developer and the Myriad 2 VPU’s capabilities, Movidius has invested heavily in developing its Myriad Development Kit (MDK). The MDK includes a graphical development environment with the tools, libraries and frameworks necessary to support efficient application development. At the core of the MDK is a code compile flow that supports C, C++ and an OpenCL front-end for maximum flexibility.
The MDK provides a groundbreaking development framework for creating arbitrary image/vision processing pipelines with automated support for data scheduling. In addition to using their own proprietary algorithms when building application pipelines, developers have access to a rich set of optimized software libraries including a computer vision library, an image signal processing library, as well as a linear algebra library. The MDK includes reference applications such as a complete Bayer camera solution, a stereo depth extraction implementation and many others. An accompanying hardware development kit comes with a full Myriad 2 reference board, support for multiple camera sensors and MEMs sensors, and an application processor integration kit.
Movidius is making Myriad 2 samples, the MDK and hardware kit available to select customers and partners under NDA in August 2014. For more information, please visit http://www.movidius.com/our-technology/myriad-2-platform. Movidius will also be presenting further details at the upcoming Hot Chips Symposium at The Flint Center in Cupertino, California, on August 12, 2014.
Movidius is a vision processor company is a vision processor company that designs compact, high-performance, ultralow power, computational imaging and vision processing chips, software, development tools and reference designs. Movidius’ architecture delivers a new wave of intelligent and contextually aware experiences for consumers in mobile and wearable devices, and other consumer and industrial camera applications. Headquartered in San Mateo, California, Movidius is a venture-backed company with investors including AIB Seed Capital Fund, Atlantic Bridge, Capital-E, DFJ Esprit and Robert Bosch Venture Capital.
Here are some info I dug out on Google’s Project Tango.
The goal of Google’s Project Tango “is to give mobile devices a human-scale understanding of space and motion.”
Project Tango devices are designed to track the full 3D motion of the device, while simultaneously creating a map of the environment. Just what does that mean in practical terms?
What if you could capture the dimensions of your home simply by walking around with your phone before you went furniture shopping? What if directions to a new location didn’t stop at the street address? What if you never again found yourself lost in a new building? What if the visually-impaired could navigate unassisted in unfamiliar indoor places? What if you could search for a product and see where the exact shelf is located in a super-store?
Imagine playing hide-and-seek in your house with your favorite game character, or transforming the hallways into a tree-lined path. Imagine competing against a friend for control over territories in your home with your own miniature army, or hiding secret virtual treasures in physical places around the world?
It’s a tall order but the technology powering “captive vision capabilities” that hope to bring awareness and perception to cameras in connected devices is by Movidious.
Google Project Tango wants to map the International Space Station:
As these vision chips continue to gain in processing power, and their prices continue to drop, we can expect to see them into our mobile cameras and the way we think about and use those cameras may undergo a drastic change.