Google has introduced a new upscaling technology that turns low-resolution images into detailed high-resolution ones It can start with a portrait as tiny as 64×64 upscale it to 1024×1024 while preserving all the detail.
Adobe Camera Raw now includes a new feature that will be coming soon to both Lightroom and Lightroom Classic. It’s called Super Resolution and it’s essentially a machine learning AI-powered method of upscaling that offers massive image upscaling benefits at the push of a button. Well, you have to push it a couple of times and check a box in a dialogue, but then it’s at the push of a button.
Julieanne Kost, Principal Evangelist at Adobe posted a video about the new feature to YouTube saying that the model for the new feature was trained on millions of photos, and uses that data to “intelligently boost the resolution of an image while maintaining clean edges and preserving important details”.
Recently, we’ve seen a bunch of upscaled and colorized historic footage: from 1911 New York to 1972 Apollo 16 Lunar Rover ride. Even videos as old as the iconic 1896 The Arrival of a Train at La Ciotat Station are possible to upscale to 4K and get a splash of color. While many of us find them inspiring and exciting, historians don’t seem to share the opinion. In fact, they argue that the whole process is “nonsense” and they’d like YouTubers to stop doing it.
NASA’s Apollo missions have left us with some iconic images and videos. One of these videos shows astronaut John Young as he was testing out the electrically powered lunar rover during the Apollo 16 mission. Denis Shiryaev took this video and gave it the same treatment as he did to a few other iconic videos. He upscaled it to 4K and colorized it, giving us an immersive, awe-inspiring experience.
I find vintage photos and videos to be something special. They give us a glimpse into the past times and tell us more about what the world was like before. But seeing them in color and high resolution can make us feel even closer to the past times, and the feeling is incredible.
This is what Denis Shiryaev did with a 1911 film A Trip to New York City. He used AI to colorize it and upscale it to 4K, so you can now see footage of New York’s daily life in color and high resolution.
AI has already been used to upscale images and increase their resolution. But how about applying it to a film? A 124-year-old silent film, to be exact? Denis Shiryaev used AI on Lumière Brothers’ The Arrival of a Train at La Ciotat Station and turned the iconic 1896 film into a 4K 60fps video.
RAISR stands for Rapid and Accurate Image Super-Resolution. It’s Google’s prototype software which utilises machine learning to provide better quality upsampling of low resolution images. They first showed off the technology in November last year, but now Google have announced that RAISR has been implemented into Google+ for Android.
The point of the technology is to save bandwidth. Many mobile users have fairly limited bandwidth. Either they have low limits, or it’s just slow. Google see RAISR as an option to save bandwidth. The idea is to scale down the images before sending out. This means they’re smaller and easier to send. Then RAISR blows them back up to their original size on the receiving end. And it wants to do this with the minimal of impacts on quality.