Using AI to recreate the past in Virtual Reality
While (too) much of the discussion around artificial intelligence or AI tends to focus on how many human jobs are going to be lost to machine-learning equipped robots in the future, we hear far less about the massive benefits that AI is going to bring to our understanding of history.
TechRadar brings us a timely feature this week to remind us how AI is soon to completely revolutionise how we map out our understanding of the past, from restoring images and paintings through to understanding far more about ancient history all the way back to the time of the dinosaurs!
Oxford University’s Deep Image Prior neural network
In terms of restoring images that have been partially eroded or damaged, there are already great developments going on with things such as Deep Image Prior, a neural network created by an Oxford University research team, and Nvidia’s image reconstruction, the graphical guru’s deep learning image reconstruction method shown off earlier this year.
“Nvidia’s process involves training its AI by taking chunks of images from image libraries ImageNet, Places2 and CelebA-HQ, which are huge repositories of images of almost every kind of common object.
“Just as you might learn to draw by sketching real-life objects, the image reconstruction algorithms here are finessed by re-drawing chunks of missing image data in these photos, then referring to the ‘complete’ original picture to see how accurate the attempt was.
“Look past the fireworks on the front end, it’s the pattern recognition in the background that makes Nvidia’s AI work like magic.”
Walking with the REAL dinosaurs
AI also tells us that “the Tyrannosaurus Rex we know from Jurassic Park bears little resemblance to what the dinosaur was actually like.
“One current theory is that such large dinosaurs may have been feathered rather than scaly, and an AI model from the University of Manchester now suggests a T-Rex couldn’t out-run a jeep either.”
University of Manchester researchers “mapped out the T-Rex’s bone and muscular structure, then used machine learning to see how fast this creature could get from point A to point B without breaking any bones.
“The findings? It’s was so big and heavy, a T-Rex could likely only walk, not run. Sprinting after some kids and scientists in search of dinner would simply put too much stress on its body.”
Perhaps most excitingly the report notes how “VR and AI are better pals than you might think. Google’s DeepMind AI lab has devised a neural network that can construct a 3D environment from as little as a single image. It extrapolates or ‘imagines’ the 3D scene based on recognition of objects and their most likely shapes. The more images it has to work with, the more faithful a replication of the real environment it can make.
“This tech has a future worth day-dreaming about. Imagine being able to recreate the home in which you grew up, in VR, using pictures from an old photo album. Or crashing your parents’ wedding as if you were a time-traveller.”