Industry Voices: Computer Vision Building Trust with AVs

Opinion piece by Eliron Ekstein, CEO and co-founder of Ravin.

In a recent pilot project, cameras, along with computer vision and artificial intelligence, at an Avis rental car station at Heathrow Airport proved to see better than humans.

In fact, the automated system found 22% more damage to returned rental cars than human workers were able to spot. This project has not made headlines the way that self-driving cars have but it is an example of how computer vision is quietly changing the entire automotive industry, not just the driving experience.

Computer vision is set to change every step of a vehicle’s life from manufacturing to sales to insurance. A look beyond self-driving vehicles, which still present safety and regulatory concerns, with their timeline constantly being pushed back, reveals many other areas of the automotive industry where computer vision is emerging, with potentially as strong, if not stronger, of an impact. Not only is computer vision already changing the way the automotive industry works but it is increasing trust and setting out a blueprint for how humans and machines can work together.

Making a difference, on and off the road

In factories, computer vision is creating a safer work environment and more uniform products. This is saving time and resources, and also improving safety for the drivers and passengers who will eventually use the vehicles. Computer vision is also behind existing advanced driving and safety features, where it is already saving lives. A study by Tsinghua University in China found that if all cars were equipped with automatic emergency braking systems, road accident fatalities would drop 13%, and injuries fall 9%. Finally, computer vision is allowing for faster and more efficient inspections of vehicles, as we saw with the Avis pilot.

Machines more objective

Computer vision throughout the industry is showing how machines are not necessarily smarter than humans, a large societal fear, but that they are more objective and that they never get tired. In the case of car-manufacturing and automated emergency braking, the reliance on a machine that never gets fatigued is obviously valuable for its potential to save lives. When used to quickly inspect rental cars for damage, or assess the cost of repairs after a vehicle is in an accident, the tireless objectivity of computer vision is again a significant benefit. Studies show that humans, regardless of the best intentions, are innately biased but artificial intelligence can help eliminate this.

Objectivity builds trust

As computer vision plays a growing role in the automotive industry, trust is likely to grow in the industry and in the technology itself. Drivers will feel safer, knowing their vehicle was manufactured without defects, and that cameras and computer vision may help them stop quickly when a child runs out in front of their car. However, when it comes to inspections, a key step of assessing damages and placing valuations on cars, the objective nature of computer vision systems will also increase trust because it takes away bias or opportunities to hide something from one party or another. It will create more trust in processes like pricing cars, buying and selling pre-owned cars and for individuals taking responsibility for damages they caused to shared or rented cars.

Trust encourages cooperation

It is important to think of these technologies not as replacing humans but helping them. This technology is allowing for the gathering and analysis of data from the physical world in order to quickly gain the type of on-demand insights we have been getting for years from the digital world. While this may not yet allow for cars to drive themselves, it does mean that in a matter of seconds, drivers can avoid hitting a pedestrian or figure out how much it will cost to fix a car after an accident.

Although many see self-driving cars as the holy grail of the automotive industry, it is important to realize that AI, and, specifically, computer vision, has a large role to play that will change the entire industry, maybe even before we actually see self-driving cars become common. It will change the whole production and selling process, as well as the user experience in everything that has to do with cars, from buying a used car to filing an insurance claim after an accident. As the project at Avis showed, machines can see better than humans and they don’t get tired. This underscores the huge potential for using them alongside humans to increase objectivity and trust in all aspects of the industry.

Leave a comment

Your email address will not be published. Required fields are marked *