The rise of artificial intelligence in recent years is grounded in the success of deep learning. Three major drivers caused the breakthrough of (deep) neural networks the availability of huge amounts of training data, powerful computational infrastructure, and advances in academia. Thereby deep learning systems start to outperform not only classical methods, but also human benchmarks in various tasks like image classification or face recognition. This creates the potential for many disruptive new businesses leveraging deep learning to solve real-world problems.
Cambridge Consultants have unveiled, what they are calling, the world's smartest car park in a move that takes deep learning out of the laboratory.
The system, which is called Goldeneye, is able to teach itself to recognize cars and how they appear in space and can do all of this without the need for expensive infrastructure.
The machine vision and deep learning solution, developed at Cambridge Consultants, uses 12 cameras to monitor 430 parking spaces and digital signs at the entrance to the site to alert the workforce and visitors to where they can quickly find a parking space.
Rather than using sensors for each individual parking space or Automatic Number Plate Recognition (ANPR) systems, Goldeneye offers a cost-effective way to scale parking monitoring via the cloud and can also be extended to include new use cases.
Beyond its deployment here it could be used to enhance wider smart city applications. Deep learning and machine vision, for example, could be harnessed to monitor traffic or to manage crowds on train platforms, or monitor crowd safety and a range of other applications.
Commenting Thomas Carmody, Head of Transport and Infrastructure at Cambridge Consultants, said "What's truly remarkable about Goldeneye is the fact that the system taught itself to identify and operate a car park. It does this without the need for any additional computing equipment."
It's a great example of a technology that's disrupting markets and with so much hype around, it's good to see deep learning being applied in the real world.