Deep Fake. Real News

by | Jun 25, 2020 | Motoring, News

Reading time: 2 minutes

Oxbotica has developed and deployed a “deepfake” technology that can generate thousands of photo-realistic images in minutes. The scalability of matching driving AI with the infinite possible situations a deepfake AI could mean a much faster development of the long-awaited Level 5 Automated Vehicle. And all without real world testing. And all without the potential risks of testing vehicles mingling with traffic.

The AV software firm believes that the pioneering technology will make the vehicles of tomorrow smarter and safer, and immediately accelerate the race to autonomy. Sophisticated deepfake algorithms allow them to try out the same scene in varied conditions. Weather, time, obstacles, and all sorts of rare events are now a click away. The technology is so advanced that it can reverse road signage, where one object (e.g. a tree) is replaced with another (e.g. a building). It can change the lighting of an image while ensuring shadows or reflections appear exactly as they should. Simulate the same object and event at different times, or seasons, or surrounded by different objects is now possible. Using synthetic images, producing thousands of true-to-life experiences and rehearsals which are not real but generated.

Oxbotica Preview

The data is generated by an advanced teaching cycle made up of two co-evolving AIs. One is attempting to create evermore convincing fake images while the other tries to detect which are real and which have been reproduced. With a feedback mechanism, these AIs try to outsmart each other, all the while improving their generation algorithms. Over time, the detection mechanism will become unable to spot the difference, which means the deepfake AI module is ready to be used to generate data to teach other AIs.

The benefit of which is not in eliminating real experiences but rather augmenting them in a way which scales arbitrarily faster than time or human resource. At any one time, Oxbotica is able to generate the experiences of any number of vehicles in any number of settings, taking into account different lighting or weather conditions.