New algorithm helps turn low-resolution images into detailed photos, ‘CSI’-style

algorithm low res high fromsmallton
Anyone who has ever worked with image files knows that, unlike the fictional world of shows like CSI, there’s no easy way to take a low-resolution image and magically transform it into a high-resolution picture using some fancy “enhance” tool. Fortunately, some brilliant computer scientists at the Max Planck Institute for Intelligent Systems in Germany are working on the problem — and they’ve come up with a pretty nifty algorithm to address it.

What they have developed is a tool called EnhanceNet-PAT, which uses artificial intelligence to create high-definition versions of low-res images. While the solution is not a miracle fix, it does produce a noticeably better result than previous attempts, thanks to some smart machine-learning algorithms.

“The task of super-resolution has been studied for decades,” Mehdi M.S. Sajjadi, one of the researchers on the project, told Digital Trends. “Before this work, even the state of the art has been producing very blurry images, especially at textured regions. The reason for this is that they asked their neural networks the impossible — to reconstruct the original image with pixel-perfect accuracy. Since this is impossible, the neural networks produce blurry results. We take a different approach [by instead asking] the neural network to produce realistic textures. To do this, the neural network takes a look at the whole image, detects regions, and uses this semantic information to produce realistic textures and sharper images.”

To train their algorithm, the researchers fed their neural network a large data set of images to build up its knowledge of different textures. The neural network only gets to see downsampled versions of the images, and is given the task of upsampling these pictures. Once the network produces an output image, the researchers then compare it with the original high-resolution image and tweak the algorithm to correct any errors, such as making it produce sharper edges or more realistic grass textures where it has not done so. After a while, the algorithm is able to do this on its own with no human intervention necessary.

“There are a lot of applications for this,” Sajjadi continued. “From upsampling old movies to 4K quality, restoring old family photographs that are too blurry when you want to get a large print, over to more general applications such as improving object detection. [It also] turns out that using our algorithm on images makes it easier for other neural networks to detect objects in images, which has wide applications, from Google image search to detecting pedestrians in self-driving cars.”

You can check out a paper describing the work here.

Mobile

Midrange phones can’t do A.I., but MediaTek’s P90 chip aims to change that

MediaTek has announced the Helio P90 mobile processor, which it says will bring the best A.I. features we see on high-end smartphones, to the mid-range. We spoke to the company about the chip.
Computing

Nvidia’s Jetson AGX Xavier module is designed to give robots better brains

Nvidia's pricey Jetson AGX Xavier might help drive the next generation of smart robots. Nvidia hopes that developers will use its new Xavier module to power AI-driven machines like delivery drones and robots used in manufacturing.
Emerging Tech

An A.I. cracks the internet’s squiggly letter bot test in 0.5 seconds

How do you prove that you’re a human when communicating on the internet? The answer used to be by solving a CAPTCHA puzzle. But maybe not for too much longer. Here is the reason why.
Movies & TV

'Prime'-time TV: Here are the best shows on Amazon Prime right now

There's more to Amazon Prime than free two-day shipping, including access to a number of phenomenal shows at no extra cost. To make the sifting easier, here are our favorite shows currently streaming on Amazon Prime.
Smart Home

This A.I.-enabled tech brings cutting-edge automation to grocery stores

Takeoff Technologies is working to make grocery deliveries fast, accurate, and convenient using A.I.-enabled technology to augment robotic grocery orders that can be completed in minutes.
Product Review

This was 3D printed? With the Anycubic Photon, you can't tell

Never mind the fact that the Anycubic Photon 3D printer only costs 500 bucks. In terms of sheer print quality, this printer is on the same level as machines that cost six times as much.
Emerging Tech

Postmates’ to roll out Minion-like autonomous delivery robots in 2019

Postmates is about to employ a cute little robot to work alongside its human delivery personnel. Called Serve, the wheel-based bot can carry items weighing up to 50 pounds and has a range of 30 miles.
Emerging Tech

The best 3D printers of 2018

On the hunt for a new 3D printer? We've got your back. Whether you're a beginner or a seasoned veteran, this list of the best 3D printers has what you're looking for.
Emerging Tech

The best drone photos from around the world

Most of today's drones come equipped with high-end cameras, which are quickly revolutionizing the world of aerial photography as we know it. Here are some of the best drone photos from around the world.
Emerging Tech

Are e-cigarettes safe? Here’s what the most recent science says

Ecigarettes are widely regarded and advertised as a healthier alternative to cigarettes for people who are trying to kick the smoking habit. How safe are these cigarette alternatives? We went deep into the recent scientific literature to…
Emerging Tech

Thrill-seekers will be able to pilot themselves in a giant drone as soon as 2019

Want to hitch a ride on a giant drone? The startup Lift Aircraft is gearing up to let paying customers fly its 18-rotor giant drones over assorted scenic landscapes across the U.S.
Emerging Tech

CRISPR gene therapy regulates hunger, staves off severe obesity in mice

Researchers from UC San Francisco have demonstrated how CRISPR gene editing can be used to prevent severe obesity in mice, without making a single edit to the mouse's genome. Here's how.
Emerging Tech

Capture app saves money by 3D scanning objects using iPhone’s TrueDepth camera

Capture is a new iPhone app created by the Y Combinator-backed startup Standard Cyborg. It allows anyone to perform 3D scans of objects and share them with buddies. Here's how it works.
Emerging Tech

Sick of walking everywhere? Here are the best electric skateboards you can buy

Thanks for Kickstarter and Indiegogo, electric skateboards are carving a bigger niche than you might think. Whether you're into speed, mileage, or something a bit more stylish, here are the best electric skateboards on the market.