Computer vision-based solutions utilize enhanced deep learning neural networks that allow data to be collected in more sophisticated ways, taking analytics to the next level: nonlinear, contextual, and accessible from multiple vantage points. Intel is leading the evolution of edge compute and computer vision solutions, helping organizations unlock new possibilities for their data with a comprehensive stacks of products designed for AI.Speaker(s):
In this session we will show how to build and deploy edge computing microservices with Eclipse ioFog. We will use OpenVINO™ on a Raspberry Pi to make some high-value microservices that perform computer vision accelerated by an Intel® Movidius™ Neural Compute Stick. Eclipse ioFog gives developers a standard way to package their software for the edge and manage it just like cloud microservices. This deep dive session will highlight the architectural approach and benefits that OpenVINO™ developers can take advantage of immediately in their work.Speaker(s):
As almost every object in your home and office become potentially internet-enabled, the IoT is poised to apply a major stress to the current internet and datacenter infrastructure. The popular approach is to centralize cloud data processing in a single site, resulting in lower costs and strong application security. However, with the sheer amount of input data that will be received from globally distributed sources, this structure will require some backup. Also, in most cases, enterprise data is pushed up to the cloud, stored and analyzed, after which a decision is made and action taken. This system is not efficient, and to make it so, there is a need to process some data in IoT case in a smart way, especially if it’s sensitive data that needs quick action.
IDC estimates the amount of data analysed on devices that are physically close to the IoT is approaching 40 percent, which supports the argument for a different approach to smart data processing. Edge/fog computing is the ideal solution to this challenge as it allows computing, decision-making and required action to happen via the IoT device itself, and pushes only relevant data to the cloud.Speaker(s):
Accelerating Time-To-Market (TTM) for your applications is critical to business success. Ensuring that you have a proven path to deploying your deep learning models in the field is therefore imperative. In this session, you will learn about the various options you have available through Intel for deployment to the edge - CPU, Integrated Graphics, Intel® Movidius™ Neural Compute Stick and FPGA. Using the Intel® Distribution of OpenVINO™ Toolkit and an Image Classification example, we will show you how to build hardware agnostic Intermediate Representations (IR) that you can then seamlessly deploy to multiple edge devices. Come join us for this exciting session!Speaker(s):
In this talk, we will cover how industrial processes are transforming from static process based operation technology to increasingly computerized industry 4.0 processes. We will discuss the concept of computer “workloads” and how they are replacing current factory processes. We will also describe how Kubernetes can be used to manage the deployment of these processes in a tested, automation pipeline.Speaker(s):
Determining if a deep learning model designed for computer vision will meet expectations when it is deployed is difficult and time consuming with existing tools and practices. Today, developers must run iterative experiments with command-line tools or custom scripting to identify throughput, latency and accuracy trade-offs. In this presentation, developer experience professionals at Intel will illustrate how a research and DX design process can be applied to create innovative deep learning solutions for computer vision. You'll have an opportunity to preview a design prototype that simplifies the process of performance profiling and tuning of deep learning models for use with Intel’s OpenVINO™ Toolkit. You'll also learn how your organization can apply DX best practices to improve customer experiences with your own products.Speaker(s):
Learn about EdgeX - the open platform for the IoT edge. In this session, Michael Estrin (Dell) and Beau Frusetta (Intel) will discuss getting started with developing on EdgeX. There will also be a demo of how EdgeX is used to run on an Intel-powered Dell Gateway, which is connected to several cameras for surveillance purposes.Speaker(s):
With both Facebook and Google's recent shift in direction towards a "Future is Private" world, learn how you too can train and deploy your AI models in a privacy-preserving way, with Decentralized AI and a combination of AI and Blockchain. These techniques will become even more rampant as we move into a world where users will own their own data and companies will start using “ethically sourced data” and move towards a path for Ethical AI for the IoT space.
In this session, you will learn:
The session will showcase insights into the drivers, solution, and benefits of real-time edge intelligence in a gas refinery, leveraging FogHorn video analytics and machine learning software, combined with ADLINK hardware.
FogHorn, a leading developer of “edge intelligence” software for industrial and commercial IoT solutions will provide a detailed overview on how leading global producers of energy and chemicals can use an Intel RRK Solution including FogHorn’s software for edge computing, real-time analytics, AI and machine learning (ML), combined with ADLINK hardware, to drive enhanced industrial performance for a variety of realworld, industrial use cases.
This session will discuss and demonstrate use cases of flare stack monitoring in a gas refinery, in combination with leading Industrial IoT Cloud solutions. The discussion will break down the challenges leading Oil & Gas companies face with flare stack monitoring, including the number of stacks monitored, limited communications, constrained compute resources, environmental and regulatory compliance goals and tightening maintenance budgets. The session will also share lessons learned to date, and the benefits edge-based analytics, AI and ML technologies offer to improve operational efficiencies, lower maintenance costs, enhance compliance-related activities and improve safety.Speaker(s):
Whether an industrial equipment company is building devices for manufacturing, energy, medical, transportation or another segment, they experience a dramatic digital transformation from a focus on hardware technology to a new modern focus on software technology that allows the consolidation of industrial systems. Two software technologies are helping to bring this digital transformation to the latest design of industrial equipment and devices: virtualization and containerization. Discover how these two technologies are enabling the consolidation of industrial systems and driving the digital transformation of the manufacturing, medical and rail transportation market segment through use cases.Speaker(s):