With a lifetime of observing the world informing our perceptions, we're all pretty good at inferring the overall shape of something we only see from the side, or for a brief moment. Computers, however, are just plain bad at it. Fortunately, a clever shortcut created by a Berkeley AI researcher may seriously improve their performance. It's useful to be able to see something in 2D and guess accurately the actual volume it takes up -- it would help with object tracking in AR and VR, creative workflows, and so on. Going up a dimension means you've got a lot more data to think about.
Researchers have figured out a way to transform a few dozen pixels into a high resolution image of a face using artificial intelligence. A team from Duke University in the US created an algorithm capable of "imagining" realistic-looking faces from blurry, unrecognisable pictures of people, with eight-times more effectiveness than previous methods. "Never have super-resolution images been created at this resolution before with this much detail," said Duke computer scientist Cynthia Rudin, who led the research. The images generated by the AI do not resemble real people, instead they are faces that look plausibly real. It therefore cannot be used to identify people from low resolution images captured by security cameras.
SEE ALSO: MashReads Podcast: How to set a New Year's reading resolution you'll actually accomplish If one of your resolutions is to write more, maybe finish that novel you've always meant to write, you're not alone. And Mashreads has your back. We went to not one, but two of YA's best authors: David Levithan and Rachel Cohn. Not only are they award-winning and bestselling authors in their own rights, but together they've collaborated on such amazing books as Nick and Norah's Infinite Playlist, Dash & Lily's Book of Dares, and its 2016 sequel The Twelve Days of Dash & Lily. Keep reading for Levithan and Cohn's advice on writing, reading, and collaborating to get you started on all those writing resolutions.
Mapping the brain is all the rage. In 2009, the National Institutes of Health announced the Human Connectome Project, an ambitious multimillion-dollar initiative to produce a detailed map of the long-range connections in the human brain. Two years later, the Allen Institute for Brain Science launched the Allen Brain Atlas, a collection of online public resources that integrate information about gene activity with neuroanatomical data. And earlier this year, President Obama announced the Brain Activity Map project, which aims to "reconstruct the full record of neural activity across complete neural circuits. Now an international team of researchers led by Katrin Amunts of the Ju lich Research Center in Germany has created the most detailed map yet of the human brain.
Scientists have created a high-resolution image of a fruit fly brain that will let researchers trace the connections of neurons throughout the brain. A team at the Howard Hughes Medical Institute's Janelia Research Campus led the work, which was recently published in Cell. Davi Bock, the lead researcher on the project, said in a statement that this level of resolution hasn't been achieved before and it will allow scientists to better understand which neurons play a role in behaviors exhibited by fruit flies. Though a fly brain is relatively small -- about the size of a poppy seed -- creating a detailed map of the 100,000 neurons it holds is still a major challenge, and traditional methods haven't allowed for this type of imaging to be done. The researchers developed a new set of tools that included high-speed cameras, custom-built systems that can quickly process brain tissue samples and a robotic loader that can pick up samples and put them into place on its own.