tech computing age software image_processing face

Software Accurately Predicts Faces As Babies Age Over Lifetime

Computer scientists can now predict how a person’s face ages from a baby to an older adult.

University of Washington researchers have developed software that automatically morphs a face through a lifetime of growing and changing no matter the lighting, expression or pose of the subject in the starting picture.

“Aging photos of very young children from a single photo is considered the most difficult of all scenarios, so we wanted to focus specifically on this very challenging case,” said study coauthor Ira Kemelmacher-Shlizerman, an assistant professor of computer science and engineering, in a university statement. “We took photos of children in completely unrestrained conditions and found that our method works remarkably well.”

See a comparison of their morphing software and the real person at different ages below.

Read More

159 notes
tech human_computer_interaction computing gesture_recognition 3_d display hologram gif

3-D Interactive Display Uses Fog As Screens

Engineers have built an interactive display using a tabletop system and mounted personal screens made of fog. Projectors light the fog for each user and a camera system monitors movements, allowing each person at the table to manipulate and share three-dimensional data.

A team at the University of Bristol in the UK say their device, called MisTable, is see-through and reach-through. Both fog screens and the table display can be manipulated by users.

"The personal screen provides direct line of sight and access to the different interaction spaces," said Sriram Subramanian, a professor of human-computer interaction. "Users can be aware of each other’s actions and can easily switch between interacting with the personal screen to the tabletop surface or the interaction section. This allows users to break in or out of shared tasks and switch between individual and group work."

Compare this to the Displair, by Russian inventor Maxim Kamanin. See the MisTable video below.

Read More

1,184 notes
science tech distributed computing citizen_science folding_home protein biology cancer health
200,000 Computers Tapped To Crack Cancer Protein

by Michael Keller

A virtual supercomputer running on more than 239,000 computers around the world has successfully eavesdropped on a protein key to cancer’s progression in the body. 

Researchers using Stanford University’s Folding@home, a distributed computational platform, have been able to describe the activation of a protein called Src kinase, a molecular switch that is believed to turn on the tumor-producing signals in cells that tell them to grow, spread and not self-destruct. 

The team says it is the first time the protein has been modeled as it changes from an inactive state to an active one. Their insight could help develop new drugs that specifically target Src kinase.

(The gif above illustrates Folding@home’s simulated protein-folding steps from an uncoiled configuration to a complex, 3-D structure. The protein here is NTL9 and unrelated to Src kinase, the subject of this article. See the interesting video below. Courtesy Vijay Pande/Stanford.)

Read More

114 notes
science tech computing neural_networks insects neurons electronics microchip
Insect Nervous System Copied To Boost Computing Power

by Charles Q. Choi

Brains are the most powerful computers known. Now microchips built to mimic insects’ nervous systems have been shown to successfully tackle technical computing problems like object recognition and data mining, researchers say.

Attempts to recreate how the brain works are nothing new. Computing principles underlying how the organ operates have inspired computer programs known as neural networks, which have been used for decades to analyze data. The artificial neurons that make up these programs imitate the brain’s neurons, with each one capable of sending, receiving and processing information.

However, real biological neural networks rely on electrical impulses known as spikes. Simulating networks of spiking neurons with software is computationally intensive, setting limits on how long these simulations can run and how large they can get.

Read More

78 notes