Latest News

How a great Visual IoT platform will power a smarter future

James Wickes, co-founder and chief executive, Cloudview, discusses the importance of visual data for smart devices

Visual data is already an important component of many aspects of our lives. It plays a central part in smart city initiatives, supports automotive developments such as self-driving vehicles, and has a place in the future of healthcare, construction, industry and many other sectors. The role visual data plays in our world is only going to grow as our appetite for smarter tech increases and our ability to apply artificial intelligence to visual data develops. The video analytics market will see a compound annual growth rate of more than 50 percent over the next five years, according to McKinsey.

With such a pivotal role in our futures, we need to be certain that visual data is open to the most advanced analytics available, and that it is stored securely. That’s why Cloudview and Digital Barriers have come together to provide the world’s first fully integrated cloud video platform. This combines ultra-low bandwidth secure live streaming with a fully scalable, resilient, easy to deploy cloud platform, in the form of EdgeVis from Digital Barriers and Cloudview’s Visual Data Platform.

It includes Digital Barriers’ government-accredited edge analytics for use with both new and existing cameras. A key aspect of the platform is its comprehensive backend management system, which ensure analytics are correctly applied and doing what they are meant to do. This is vital to producing good quality, meaningful insights. The old computing adage ‘garbage in, garbage out’ applies as much today as it ever did.

Visual data applications require a number of specific features that are expensive and technologically challenging for organisations to develop for themselves. A start-up with some great ideas for using visual data to better protect vulnerable people such as helping dementia patients live independently in their own homes, or identifying potentially suicidal people at railway stations in time for intervention to save their lives, should not have to face the costly, time consuming and technologically challenging barriers involved in developing a platform to power their idea. They should be able to focus on the practicalities.

This start-up will need the ability to stream high quality visual data constantly into a platform which can apply analytics and deliver information to individuals that is of a high enough quality to enable them to make crucial decisions. Is the person on the railway platform behaving in a way which causes concern? Should an intervention be triggered? Does the behaviour pattern of a person in their own home, that’s been flagged as concerning by the AI, warrant action, and if so what action should be taken?

A platform that can provide the granular information needed for people to make such decisions will need to lever both cloud and edge computing to handle data quickly enough to deliver what’s required. The visual data itself, and other metadata relating to it, about individuals, may well be sensitive and will need to be compliant with data protection and personal privacy regulations.

The analytics components required by a Visual IoT service such as our hypothetical examples might be in part ‘off the shelf’ and in part bespoke for the particular service that’s envisioned. Examples of ‘off the shelf’ components include number plate recognition, facial recognition, behavioural analysis, intrusion detection, and analysis of the flow of vehicles or people through particular spaces. Being able to access such features from a platform, almost in a sort of ‘pick-and-mix’ way, alongside having access to developers who can build bespoke elements, will allow developers to work on complex projects without needing to worry about where to source the analytics components they need. They will be able to prototype and test ideas, and bring the best versions of their work to market at speed.

This is why our platform is entirely API driven. Both new and yet to be created analytics and extensions can be developed to work with existing APIs, so the system will be as open as possible to the requirements of Visual IoT projects whose range and scope we can’t possibly predict.

Our ambition is to open the platform up to a broader range of Visual IoT solutions so that our hypothetical developer that wants to help vulnerable people, and many thousands of others, can do what they want to do without worrying about streaming technologies, analytics components and legal compliance. Smart start-ups with great ideas really don’t need the hassle or bottlenecks such worries create for them. And, of course, established players in Visual IoT can also take advantage, using the platform to build on their existing experience and knowledge at a greater speed.

For further information visit