6 Ideas to Use AI for Life-Like Virtual & Augmented Reality

Sophia Brooke
Virtual Reality Pop
5 min readOct 1, 2018

--

A few decades ago, any vision of the future included some form of Augmented Reality (AR) or Virtual Reality (VR). Today, this goal is still under construction, as it has to overcome significant technical issues related to rendering and optimization. These issues aren’t only related to highly pretentious users who prefer crystal clear views on their Oculus. Physiological reactions which are similar to sea-sickness need to be prevented and it’s important to note that currently, VR devices can’t be used for more than 20 minutes.

Until now, all advancements focused on better hardware and mainly an increased and uninterrupted frame-rate. Only recently has the idea to using artificial intelligence software development for AR/VR emerged, this will make it a lot more credible and also cut costs.

The reasoning behind this is simple: AI and Big Data are ideally suited for pattern recognition and thus, similar pattern generation. This way of working can replace whole teams of graphic artists and generate new environments starting from a few pictures. Let’s see some other capabilities of this idea and some real-world applications.

Simulators

The first application we can think about for VR is related to simulators of all kinds, from entertaining ones to educational or military. You could learn to drive a car, a motorcycle or a plane, safely, without endangering your life or that of the instructor. Also, for some activities, this would mean an important diminishing of costs, as practice in some areas involves expensive supplies, like kerosene.

AI could replicate numerous situations which occur randomly and learn from the behavior of the student. As the trainee gets better, through a gamification approach, the system will present ever-increasing difficult conditions.

Not only could VR train our future drivers and pilots, but our future surgeons too. You can rest assured that such a healthcare professional would have seen more cases than well-versed physicians, and that is even before finishing school. This will happen due to the system’s vast library of cases presented to the user.

Environment mapping

Ikea and other furniture providers have apps which offer users the opportunity to try-out furniture in their rooms, after carefully inputting the size and potential obstacles like doors and windows. Now imagine this process being faster and more accurate by simply scanning your room with your phone.

AI could not only scan the room and create a realistic version where the user can place items as they desire. With a bit more training, it could offer on-demand design services. For example, the user will select a style, the necessary things like a bed, a desk, or maybe a nightstand and the program will come up with a complete scheme, much like an interior designer would do. By refining this initial proposition, the user will create the final design by allowing the AI to learn as it goes.

Early warnings

Faced with the problem of survival, humans have developed mechanisms to keep them safe from danger and the ability to judge a situation in less than a second. However, these judgments which we usually call intuition are not infallible. Imagine if such a task could be overtaken by a machine which has the combined experience of thousands of men.

Such a creation could save millions on the battlefield by anticipating the opponent’s moves and alerting soldiers in advance.

On top of that, by using the previously described application, the AI could map the battlefield in an AR style, placing warning signs. This sounds like turning war into a video game and people into cyborgs, but it’s a matter of survival.

Game development

We’ve already mentioned gaming as a primary application. It is safe to assume it will continue to be one of the driving forces and financing sources for AR’s and VR’s progression. AI can become a noteworthy assistant in this endeavor.

First, it will replace the current method of animation. AI will learn the possible movements of the character as well as how these can combine and create realistic outcomes on any type of new character, simply adjusting to the shape. The advantage is that it can learn by demonstration, but it is also self-teaching, which means that it can even come up with new moves or performances, adding to the excitement of the game.

AI could transform the gaming environment entirely and make even non-player characters part of the story in a more relatable and believable way. This can be achieved by studying footage retrieved from CCTV and replicating the reactions of people.

Immersive experiences: Traveling and social media

Virtual reality is not only about fantastic worlds where we can lose ourselves. It can also be an excellent replica of places in the real world which are too far away and costly to reach. Creating immersive travel experiences could be as close as it gets to the real thing for some demographics like the disabled or elderly. It could become a new form of entertainment for those who have the travel bug and want to replace Netflix with the Himalayas.

Even social media will get a boost from the VR immersive experiences powered by AI. Facebook’s acquisition of Oculus Rift is a hint about this new direction. AI will have the task of designing your social media avatar taking into consideration both your pictures and your preferences, a technique already used in client profiling.

Risk management

Sci-Fi movies and literature might have inspired the application of AI for AR. Now it is used to create instant profiling of potential clients, employees, and even romantic partners. Everyone would appreciate having a pair of magical glasses which show information about their interlocutor and warn about misconduct. Although this technique sounds highly biased or intrusive, it is not such a far cry from what is currently happening when you apply for a loan.

These are only a handful of the possible applications of AI in AR and VR. As the technology develops and it becomes more widely accepted there will be more ways to use it even for trivial matters like selecting clothes to high-end uses including building design.

--

--