Wednesday November 13, 2019
Home Lead Story Adobe Built a...

Adobe Built an AI To Spot Photoshopped Faces

This new research is part of a broader effort across Adobe to better detect image, video, audio and document manipulations

0
//
Adobe, AI, Photoshopped
The decision is also intended to make image forensics understandable for everyone. Pixabay

Adobe, along with researchers from the University of California, Berkeley, have trained Artificial Intelligence (AI) to detect facial manipulation in images edited using the Photoshop software.

At a time when deepfake visual content is getting commoner and more deceptive, the decision is also intended to make image forensics understandable for everyone.

“This new research is part of a broader effort across Adobe to better detect image, video, audio and document manipulations,” the company wrote in a blog-post on Friday.

As part of the programme, the team trained a convolutional neural network (CNN) to spot changes in images made with Photoshop’s “Face Away Liquify” feature, which was intentionally designed to change facial features like eyes and mouth.

Adobe, AI, Photoshopped
Adobe, along with researchers from the University of California, Berkeley, have trained Artificial Intelligence (AI) to detect facial manipulation in images edited using the Photoshop software. Pixabay

On testing, it was found that while human eyes were able to judge the altered face 53 per cent of the time, the the trained neural network tool achieved results as high as 99 per cent.

The tool also identified specific areas and methods of facial warping.

Adobe’s execution in detecting facial manipulation came just days after doctored videos of Facebook CEO Mark Zuckerberg and US Speaker Nancy Pelosi made the rounds on social media as well as news channels.

“This is an important step in being able to detect certain types of image editing, and the undo capability works surprisingly well. Beyond technologies like this, the best defence will be a sophisticated public who know that content can be manipulated, often to delight them, but sometimes to mislead them as well,” said Gavin Miller, Head of Research, Adobe.

Adobe’s Photoshop software was originally released in 1990. (IANS)

Next Story

New Device Can Detect Unattended Kids Or Animals

The new device developed in University of Waterloo, Canada can detect unattended kids and animals

0
device
This new device can help save kids and animals in vehicles. Pixabay

Researchers from University of Waterloo, Canada, have developed a new device which combines radar technology with Artificial Intelligence (AI) to detect unattended children or animals with a 100 per cent accuracy.

Small enough to fit in the palm of a hand at just 3nm in diameter, the device is designed to be attached to a vehicle’s rear-view mirror or mounted on the ceiling.

According to the researchers, it sends out radar signals that are reflected back by people, animals and objects in the vehicle. The built-in AI then analyses the reflected signals.

“Unlike cameras, this device preserves privacy and it doesn’t have any blind spots because radar can penetrate seats, for instance, to determine if there is an infant in a rear-facing car seat,” said study researcher George Shaker, a Professor at the University.

“This system is so affordable it could become standard equipment in all vehicles,” he added.

Development of the wireless, disc-shaped sensor was funded in part by a major automotive parts manufacturer that is aiming to bring it to market by the end of 2020.

Analysis by the device determines the number of occupants and their locations in a vehicle.

That information could be used to set rates for ride-sharing services and toll roads, or to qualify vehicles for car-pool lanes.

device
According to the researchers, the device sends out radar signals that are reflected back by people, animals and objects in the vehicle. Pixabay

Its primary purpose, however, is to detect when a child or pet has been accidentally or deliberately left behind, a scenario that can result in serious harm or death in extremely hot or cold weather, the study said.

In such cases, the system would prevent vehicle doors from locking and sound an alarm to alert the driver, passengers and other people in the area that there is a problem.

The low-power device, which runs on a vehicle’s battery, distinguishes between living beings and inanimate objects by detecting subtle breathing movements.

Also Read- Xiaomi’s Foldable Phone may Come with Five Pop-Up Cameras

Researchers were now also exploring the use of that capability to monitor the vital signs of drivers for indications of fatigue, distraction, impairment, illness or other issues.

The study was presented at the IEEE Sensors 2019 conference in Canada. (IANS)