Neural Photo Editor works like magic thanks to machine learning

By Daven Mathies September 26, 2016

Neural Photo Editing with Introspective Adversarial Networks

Neural Photo Editor is an experimental piece of retouching software from researchers at the University of Edinburgh that uses neural networks to act like Photoshop on steroids. Thanks to machine learning, it can intuitively interpret how a user intends to retouch a photo based on a “contextual paintbrush.” A single brush can change hair color, fill in bald spots, or add a toothy grin.

The process couldn’t be simpler: Users select a color for their paintbrush and the system analyzes that color in context with the image in order to produce an intelligent output. Painting over a subject’s mouth with a white brush, for example, can make a smile larger, while painting with a dark color on a forehead can add bangs.

The software currently works best on images generated by it that have constrained limits on what can be manipulated. However, through the use of “introspective adversarial networks,” an advanced masking system allows it to be applied to existing images, as well.

The tech-savvy can download Neural Photo Editor from Github, but don’t go canceling your Adobe Creative Cloud subscription anytime soon. The tool is very much in its early days and is far from perfect. While it produces stable outcomes much of the time, it can still lead to bizarre results on occasion, as the video above demonstrates. The video also shows the system working with very low-resolution images, so the practical applications for such a tool for photographers are currently nonexistent.

Still, it is an intriguing look at how photo editing could evolve in the future to be smarter and more context-aware, significantly increasing the speed at which a photo could be majorly retouched. It also represents a practical application for the general use of neural networks in personal computing.

Editors' Recommendations

Topics

Former Digital Trends Contributor

Daven is a contributing writer to the photography section. He has been with Digital Trends since 2016 and has been writing…

Photography

Adobe and UC Berkeley team up to develop neural network image editor

adobe uc berkeley neural image editing network

Machine learning is all the rage these days, and it could have a future in Photoshop. Adobe is working with researchers at UC Berkeley to develop software that uses artificial intelligence to help users realistically manipulate objects in an image with a few simple brush strokes.

Like the Neural Photo Editor project from researchers at the University of Edinburgh we covered earlier this week, the Adobe/Berkeley project uses adversarial networks to train the system to make realistic edits based on minimum user input. Three brush tools -- coloring, sketching, and warping -- allow for different aspects of the image to be manipulated. The demonstration video above shows how a user can simply paint a stroke of blue over a black shoe and the system automatically applies a blue shade to the entire shoe.

Web

Trust issues? Use Trooly and machine learning to figure out who you may be working with

trooly instant trust rating handshake shutterstock

You may think you're a great judge of character, but how much do you really know about someone after a single meeting? Despite the importance of trust in any relationship, business or otherwise, it can be hard to ascertain, especially in a short period of time. So to help you make better, more informed decisions about people you may want to work with, tech company Trooly has launched its Instant Trust rating service, which claims to help "businesses verify, screen, and predict trustworthy relationships and interactions." It's all based on machine learning and the wealth of information available within your digital footprint, and Trooly says it hopes to fill the "trust gap" that results from the "speed of modern commerce and community."

Available to both businesses and consumers, Trooly uses data that is generally publicly available to better understand an individual's -- or a company's -- personality and behavior traits. These characteristics, the company says, can be used to predict trust and safety-management applications, risk applications, customer relationship management, and more. And on the individual level, Trooly's machine learning score claims to reflect a person's strengths and trustworthiness -- you can even look into your own score to see how you appear to others (and maybe work on cleaning things up, should it be necessary).

Computing

Fathom Neural Compute Stick makes deep learning possible for $100

movidius neural networking chip under 100 dollars fathom

Remember being impressed that an entire computer could fit onto a mere USB stick? That appears to already be old hat, as chip maker Movidius has announced the Fathom Neural Compute Stick, a new device that puts the power of deep learning in your pocket.

Movidius has been making some waves in this field recently. In January, the startup inked a deal with Google to assist with its efforts to improve image recognition on mobile devices, and its impressive 'visual processing unit' intended for use in VR headsets garnered some buzz just a few weeks earlier.