CASE STUDY：
Deploying an offline trained model to Xvisio technology’s SlimEdge XR Starter Kit seamlessly and effectively
Augmented Reality (AR) is continuously growing and capturing people’s attention. Potential uses and applications of AR are evolving rapidly across a wide range of industry verticals, such as entertainment, education, retail, healthcare, transportation and manufacturing. New AR technologies continue to garner widespread consumer and enterprise interest, with the value of the global AR/VR market expected to touch $209.2 billion by 2022.1 The explosive growth of the segment can be attributed to the ubiquity of AR applications since AR can potentially be leveraged anywhere, becoming a common presence at work or in everyday life.
Hardware components required for AR are a processor, display, sensors, and input devices. AR devices are regarded as the next generation computing platform, especially with the imminent deployment of 5G networks. It is becoming increasingly evident that smartphones, AR glasses and headsets will coexist and expand significantly the range of possible AR experiences of tomorrow.
Challenge
Though AR solutions continue to witness increased demand, developers in the sector still face several challenges. The complexity of building an AR system requires advanced spatial sensing and computing technology based on vSLAM – a computing intensive challenging task for wearable devices. Also, the AR device is a typical IoT-based hardware that requires local data collection and processing to enhance user experience. As a new computing platform and tool, AI support is an essential capability in such devices. Hence, the two biggest questions facing developers are:
• How can you enable users to develop content easily?
• How to leverage the massive models trained offline using different frameworks and tools?
Solution
Xvisio Technology has risen to the challenges faced by AR solution providers by deploying the vSLAM engine, depth sensing engine, video and audio processing engine, display unit, and the convolutional neural networks (CNN) based inferencing engine on a single platform. The SlimEdge XR Starter Kit B50, powered by Intel® Movidius™ VPUs, is a well-calibrated device with multiple cameras and sensors for AR, AI, and computer vision (CV). It offers a user-friendly environment to implement offline trained models to the SlimEdge XR, using the Intel® Distribution of OpenVINO™ toolkit. The kit also empowers developers to seamlessly deploy AR applications with state-of-the-art AI technology.
About the Product: SlimEdge XR Starter Kit
Leveraging the Intel® Distribution of OpenVINO™ Toolkit
Collaborating with Intel, Xvisio is incorporating deep learning and inferencing capabilities of the Intel® Distribution of OpenVINO™ toolkit to deploy an offline trained model to SlimEdge XR Starter Kit. OpenVINO™ toolkit is part of Xvisio’s software tools for network calibration and implementation. Xvisio uses Intel® Movidius™ VPUs to handle edge inferencing tasks for AR glasses and robotics applications. Xvisio offers platforms that enable users to develop their own applications. OpenCV and OpenVINO™ toolkit included in the tool chain help leverage vast amount of existing resources to accelerate development.
Conclusion
AR devices, empowered by state-ofthe-art AI technology, are changing the way we interact with the world, and are gradually becoming the next generation personal device and work equipment. However, there are still several roadblocks facing AR developers, especially in deploying edge inferencing capabilities in devices. Working together with Intel, Xvisio has created a successful solution to enable fast implementation of edge inferencing capability for AR platforms with the Intel® Movidius™ VPU family and the Intel® Distribution of OpenVINO™ toolkit.
About Intel® Distribution of OpenVINO™ Toolkit
The Intel® Distribution of OpenVINO™ toolkit, an open visual inference and neural network optimization toolkit, is free software that helps developers and data scientists speed up computer vision workloads, streamline deep learning inference and deployments, and enable easy, heterogeneous execution across Intel’s platforms – from edge to cloud.
1Source: https://techjury.net/blog/virtual-reality-statistics/
Intel technologies may require enabled hardware, software or service activation.
Intel’s compilers may or may not optimize to the same degree for non-Intel microprocessors for optimizations that are not unique to Intel microprocessors. These optimizations include SSE2, SSE3, and SSSE3 instruction sets and other optimizations. Intel does not guarantee the availability, functionality, or effectiveness of any optimization on microprocessors not manufactured by Intel. Microprocessor-dependent optimizations in this product are intended for use with Intel microprocessors. Certain optimizations not specific to Intel microarchitecture are reserved for Intel microprocessors. Please refer to the applicable product User and Reference Guides for more information regarding the specific instruction sets covered by this notice.
Intel disclaims all express and implied warranties, including without limitation, the implied warranties of merchantability, fitness for a particular purpose, and non-infringement, as well as any warranty arising from course of performance, course of dealing, or usage in trade.
Intel does not control or audit third-party data. You should consult other sources to evaluate accuracy.
Intel is committed to respecting human rights and avoiding complicity in human rights abuses. See Intel’s Global Human Rights Principles. Intel’s products and software are intended only to be used in applications that do not cause or contribute to a violation of an internationally recognized human right.
No product or component can be absolutely secure.
Your costs and results may vary.
© Intel Corporation. Intel, the Intel logo, and other Intel marks are trademarks of Intel Corporation or its subsidiaries.
Other names and brands may be claimed as the property of others.
(SOURCE：INTEL)
About Xvisio Technology
Xvisio Technology was founded in Silicon Valley in 2016 and now has expanded its presence to China and Europe. As the leading solution provider for spatial perception interaction, it provides innovative, fundamental technologies and integrated solutions that power AR/VR glasses.
SlimEdge XR is an integrated edge platform that combines VSLAM engine, depth engine, AI engine as well as hand gesture and eye tracking engine into one system. It’s well suited for USB tethered or all in one AR/VR glass or HMDs.