In the world of Internet of Things, it is well known that the critical mass of digital data resides and is collected at the node or network edge. Bottlenecks to omnipresent AI, such as power consumption, bandwidth, latency, network connectivity, security and privacy have led to recent innovations in on-device edge computing for AI.
Gartner projects that IoT devices will climb to a total of 50 billion by 2020 and the market is expected to grow from USD 185.8 Million in 2017 to USD 838.6 Million by 2022, at a Compound Annual Growth Rate (CAGR) of 35.2%, unleashing a wave of services and applications at the network edge that can be optimized with artificial intelligence.
Today, in most cases, IoT and AI work together in cloud computing. Data from IoT devices is transmitted back to a central hub in the cloud where it is analyzed and stored and actionable insights are sent back to the device. In this session, we will discuss about some real-life use-cases of performing AI on edge and its importance; followed by the challenges of the presently used approaches in terms of latency, network connectivity, privacy, cost, etc and its possible solutions. Few of the use cases for edge AI includes smart security cameras, UAVs, ADAS, health-tech, etc.
We will discuss the ways to move existing AI applications from the cloud to the edge and will be showcasing demos of deploying edge applications using Intel Movidius Neural Compute Stick (NCS) which is a tiny, fan-less, low-cost deep learning device that can be used to perform AI at the edge. We will also provide a brief overview of OpenVINO – 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® platforms from edge to cloud.
Mashrin Srivastava is part of the Data Centre Engineering & Architecture team in Data Centre Group at Intel, specializing on Intel platforms and products and has worked on projects in Data Science/ IoT space including connected homes, smart city, mobility and edge AI. He has exhibited his contributions and represented Intel at several conferences.
He has a Bachelor’s Degree in Computer Science and Engineering from Vellore Institute of Technology. He is passionate about data science, algorithms and graph and the use of the technology to bring disruptive changes. He has served as a contributor and Directly Responsible Individual to the Stanford Scholar Initiative and open source contributor to Processing Foundation.
Krishnakumar Shetti (Krishna) is a Technical Consulting Engineer in the Intel Core and Visual Computing Group (CVCG), specializing in Intel products and tools related to Embedded/IoT space.
His experience includes system debug, power/performance analysis and edge analytics on various Intel® platforms. He has a Master’s Degree in Embedded Systems from Manipal University.