The Uncanny Valley is an interesting concept, first identified in the 1970’s, whereby a human experiences a realistic human-like simulation, robot or animation and is repulsed by the subtle differences that seem false or unreal, which the brain is easily able to pick up on. The end result being that the human brain is unsettled and uncomfortable, even if the technology behind that particular example is very advanced — hardly ideal when the main goal of many of these projects is to entertain and enrich lives.
The reason for this reaction is that humans have evolved to be able to read micro-expressions, body language and other subtle nuances of human form. The animation or android triggers activity in the part of the brain which evaluates and monitors errors in movement — perceiving the way the replica moves and gestures as strange anomalies, making it almost untrustworthy and definitely unsettling.
This phenomenon has been at the crux of many challenges film producers and games developers face when trying to portray realistic and lifelike characters using CGI. Pre noughties — films like Michael Crichton’s Westworld, James Cameron’s Terminator and Steven Lisberger’s Tron included some of the earliest CGI experiments. Some of the lengths they had to go to (including limitations such as working with a computer with just 2Mb of memory, unheard of in current times) serve as testament to how far computing and technology has advanced since then, and what can be accomplished with contemporary tech.
The gaming industry has also had a difficult time accurately portraying characters that are supposed to be relatable physically. In 2004 Valve released Half-Life 2 and although the graphics were incredible for the time and certainly moved gaming away from the Uncanny Valley, it was reported that they didn’t quite hit the realism needed for a person to be unaware of the fact the images were computer generated. This still remains a challenge — but not a complete roadblock. Countless advancements in 3D modelling, dynamic lighting and processing power available to the general public, means that this issue is eroding over time, but to date this issue can pervade in even the latest game releases boasting the best graphics.
Some film producers such as Pixar avoid mimicking real-life proportions altogether, which leads the brain to feel more at ease with the pseudo realism. Although many of the Pixar characters we are familiar with are human-like, they remain entirely separate from us due to their cartoonized stylization. They do exhibit some humanized expression, but the animators don’t try to emulate humans fully, so we perceive it on an entirely different level.
Virtual Reality advancements will be next in line to tackle this challenge, but with the added considerations for the medium; VR headsets are still evolving. Current tech is vastly different from big-screens of old. VR headsets at this point in time suffer from screen door effect — a rather un-immersive side effect where the viewer can see the pixels of the image very close to your eyes, due to two separate screens being driven at a high resolution at the same time.
As with each piece of technology designed to mimic realism thus far, it’s a case of time, patience and refinement. Computing power is continually increasing. Techniques and processes are improving in the right direction. Most recently, an incredibly realistic CGI character was revealed to the world, built by a couple from Tokyo — which left some viewers astonished at the accuracy ‘Saya’ displays in the digital rendering of human features.
Not long after the reveal of Saya, artists, actors and developers collaborated to produce a very recently released a real-time performance AI named Siren, based on the likeness of Chinese actress Bingjie Jang. At a recent Games Development Conference, attendees could actually interact with Siren live, with actress Alexa Lee driving the interaction with a motion capture powered performance.
The continual development and pioneering tech surrounding virtual realism is definitely promising and clearly has big corporate support. With these continual advancements in mind, it’s not too much of a stretch to imagine that one day very soon, we could be interacting with computer-generated humans on a daily basis without blinking an eye. Developers are creating more and more convincing animations to stand at the centre of gaming and VR experiences, but that isn’t necessarily where it ends — one day simulated human beings could walk among us and the ‘Uncanny Valley’ will be a thing of the past.
Ciaran Foley is CEO of Ukledo and Immersive Entertainment, Inc. a Southern California virtual reality software company developing a new virtual engagement platform called Virtual Universe (VU).
___________________________________________________________________
Learn more about Virtual Universe and VU token by visiting our website and signing up for email updates, visiting our Github, following us on Twitter, Facebook, Linkedin, and Instagram, or being part the discussion on Telegram and Discord.
