Video is an information-rich, highly flexible information source - so much so that humans can find it immersive. We can analyse video and image based input from any source, including: infrared, visible light, and x-ray based input sources. We can process live streams or pre-recorded content, in low resolution or high resolution. Hence we can help you to achieve a greater return on investment for existing infrastructure, or a greater return on investment in new projects.
That richness enables video to capture information pertinent to just about any business. For example, you might want to ensure that:
• your processes are running smoothly
• your customers are receiving good service
• available assets are being utilised and optimised
• potential risks are spotted quickly
• your customers, employees & assets are safe
Having people analyse video all day isn't viable for most applications. Even where it is viable, applications will benefit from automation - to enable people to focus on the priorities and to have 'extra eyes' as a fallback.
To respond quickly, we need to automate with computers to provide assistive intelligence.
The richness which makes video so flexible, also makes processing video for more detailed analysis a real headache for computers.
There are a few exceptions. In some cases you don't really need to know what is in a video. If detecting motion, or basic objects within a zone is enough, then simple methods can now be deployed within commodity cameras. However, most video analysis stretches even sophisticated techniques.
The prospect of self-driving cars, whether 3 years away or 10 years away, enables us to imagine what will be possible when computers can assess the information available within video sources to enable all kinds of automation.
How is this legacy? Because even these systems based on deep learning and derivative algorithms are fundamentally based on ideas that are now 30 years old. Their achievements come largely thanks to massive data processing power. These approaches are brittle, even after billions of dollars of investment to pay lots of very smart people to design and train with huge quantities of data.
Whatever your business requirements, the principle of specialist people acquiring lots of data and the costs associated with that still hold. For custom applications you would bear these costs. For pre-built applications these costs have to be passed on to you.
Our technology looks at video very differently. We don't use legacy deep learning approaches. Our approach makes assisting you much simpler. We can train very rapidly, to deliver highly accurate systems with flexibility. This allows for a much broader set of requirements to be achieved, with greater agility to deliver sooner and respond to change more rapidly.
• are an innovator in your industry
• believe that appropriately digitised video input sources (infrared, light, xray) could provide real value to your business, to solve real challenges
• want to work with true innovation from a highly specialised tech startup
• have a modest amount of time and budget available to allow us to test what is possible together
Our office is located at the Malvern Hills Science Park, in the UK.
Sidaway Technologies Ltd, Malvern Hills Science Park, Unit 50-51, Geraldine Road, Great Malvern, WR14 3SZ.