AnyVision Fast-Tracks Development of AI-based Vision Applications

Using Intel® DevCloud for the Edge as a sandbox for testing, prototyping, and solution optimization, AnyVision fine tunes its edge-based computer vision and AI solutions for a growing need for visual intelligence in security and payment systems.

 

Article Key

Image credit: AnyVIsion: Insights Smart Space Optimization Platform

Living through a pandemic is changing the way we behave. With widespread concern about the spread of infection, people are seeking ways to interact with fewer touchpoints. Advanced computer technologies and the Internet of Things is making it easier to accomplish this.

The magic behind the technology is artificial intelligence, computer vision, and data analytics, and to be successful those require highly efficient computer processing capabilities and intuitive systems. As companies restructure and adjust to this new world, they need system integrators (SIs) who can build customized solutions in an efficient and cost-effective manner.

One study found that the AI in computer vision market is estimated to reach $25.32 billion by 2023, at an CAGR or 47.54 percent over the next few years. Driving that growth is increasing demand for computer vision systems in emerging and non-traditional applications and for edge computing in mobile devices.

Computer Vision and AI

AnyVision is helping to lead the charge. The company develops body and object recognition software, using deep learning algorithms, real-time analytics, and authentication processes. It provides solutions for security/building entry, retailers, smart homes, smart cities, and a variety of other applications. Heightened concerns about public health and COVID-19 transmission are pushing innovations in secure, touchless access to buildings and transportation centers, as well as contact tracing with alerts that ensure privacy. 

As employees work from home, AnyVision technologies can authenticate and verify user access on personal devices to ensure network security and prevent fraud. The company has also created a mini-box edge device to bring computer vision and deep learning capabilities to a client’s existing cameras and edge devices.

Processing Vast Amounts of Data

Because computer vision involves processing and analyzing vast amounts of images, edge computing is critical to accelerate processing and avoid network latency. Helping SIs navigate this is Intel® DevCloud for the Edge, a cloud-based development environment that enables companies to design, build, and test their products’ capabilities. Developers can analyze performance and tweak solutions using Intel’s hardware and software, with access to Intel’s optimized frameworks, tools, and libraries.   

AnyVision tested its algorithms using Intel® DevCloud for the Edge to benchmark latency and performance on 2nd Generation Intel® Xeon® Scalable processors. Performance was twice as fast when the team applied Intel® Deep Learning Boost, a platform feature specially designed to enhance inference. AnyVision also used the Intel® Distribution of OpenVINO™ toolkit, which enables the company to adapt the solution across other types of hardware for multiple clients.

“Because we serve customers with so many different needs, it’s important to quickly achieve the right balance of price and performance for each of our applications. Intel® DevCloud for the Edge lets us test multiple platforms in parallel. That’s a lot of time savings—and time is money—so it’s a no-brainer,” says Eduard Vazquez, research technical manager with AnyVision.

Finding the Sweet Spot for Performance and Precision

Intel® DevCloud for the Edge allows solution integrators to prototype, build, and optimize cloud-hosted solutions, using Intel optimized hardware and software. SIs and developers who join the community can use Intel® DevCloud for the Edge to refine their solutions, innovate, and speed time to market. The key components include:

  • The integrated Intel® Distribution of OpenVINO™ toolkit provides a flexible software development environment for deep learning inference.
  • For performance benchmarking, developers and SIs receive hyper-customized benchmarks in an easy-to-compare, side-by-side format.
  • Intel® DevCloud for the Edge contains tutorials and examples in a cloud-native developer environment to help SIs who may not know where to get started.
  • SIs get 30 days of access (with available extensions) and 50 GB of file storage.

Learn, build, and optimize with computer vision solutions. Sign up for Intel® DevCloud for the Edge at devcloud.intel.com/edge

Note: 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.