Will AWS Panorama bring new dimension to the computer vision game?

AWS re:Invent 2020, has been one of the most anticipated cloud computing events this year. Organized by Amazon, this event is available virtually for a free three-week, from Nov. 30 to Dec. 18, during which many new AWS cloud computing products are unveiled. Under machine learning based services, the cloud giant has launched five services viz., Amazon Monitron, Amazon Lookout for Equipment, the AWS Panorama Appliance, the AWS Panorama SDK, and Amazon Lookout for Vision.

Amazon says that these new machine learning services are poised to help industrial and manufacturing companies embed intelligence in their production processes for improving operational efficiency, quality control, security, and workplace safety. These services are built on avant grade machine learning, sensor analysis, and computer vision capabilities. The key objective of these services is to tackle generic technical challenges encountered by industrial customers and represent the most comprehensive suite of cloud-to-edge industrial machine learning services available. Out of these machine learning AWS services, AWS Panorama Appliance and AWS Panorama SDK are receiving huge spotlight.

According to Amazon blog post, the AWS Panorama package is a machine learning Appliance and Software Development Kit (SDK) focused to aid companies to bring computer vision (CV) to on-premises cameras to make predictions locally with high accuracy and low latency. The AWS Panorama Appliance is a hardware device that allows users to add computer to the internet protocol (IP) cameras that weren't built to accommodate computer vision but may have already been deployed. Whereas, AWS Panorama Device SDK is a software kit that enables third-party industrial camera manufacturers to build new cameras embedded with meaningful computer vision models at the edge for tasks like object detection or activity recognition. The cameras built using AWS Panorama Device SDK, have ability for processing higher quality video with better resolution for spotting issues. Also these cameras can use chips designed for computer vision from NVIDIA Jetson product family and Ambarella CV 2x product line.

Meanwhile, AWS Panorama Applicance can turn existing cameras into smart cameras that can run computer vision models on multiple concurrent video streams and IP cameras. Currently, Panorama Appliance isn't available except for preview in US East (N. Virginia) and US West (Oregon) regions. However, a user can develop a computer vision model using Amazon SageMaker and then deploy it to a Panorama Appliance to run the model on video feeds. Users can also train their own models in Amazon SageMaker and deploy them on cameras built with the AWS Panorama SDK. Customers can also enhance the ability of cameras built with the AWS Panorama SDK by adding Lambda functions to be alerted about potential issues via text or email.

Today, computer vision is heralded as most popular application of artificial intelligence. This application of AI understands visual data by mirroring functions of human eye.  Its immense potential empower a range of industries, from manufacturing to retail to finance, helping businesses extend and enhance AI at the edge.

As AWS Panorama adds computer vision functionalities to camera systems, it can analyze video feeds within milliseconds from a single management interface across a variety of locations, thus facilitating real-time visibility into operations and fast decision making. It also lets users have accessibility to real-time predictions in area where network connectivity is slow, expensive, or intermittent.

Amazon states that AWS Panorama can identify and resolve issues faster to help improve product quality and decrease costs in a manufacturing company setting. It can also discover defects and issue alerts for immediate corrective action. In supply chain network, leveraging AWS Panorama can track throughput, recognize bar codes or labels of parts, and optimize freight operations in transport yard. Furthermore, in retail sector, it can augment operations and customer experience by providing real-time insights.