figurer scientists at the University of Washington ( UW ) have witness a means to make everyHarry Potterfan ’s dreaming come straight by get two - dimensional photos to life . ( Sorry kids , the magic is in the technology and not the wizarding populace . )
An algorithm calledPhoto Wake - Upuses augment reality tools to fetch animated movement to an object from a two - dimensional photo or small-arm of artwork by contribute them the power to walk , run , and even jump .
“ This is a very concentrated underlying problem in reckoner vision , ” said study co - authorIra Kemelmacher - Shlizermanin astatement . “ The big challenge here is that the input signal is only from a single tv camera situation , so part of the mortal is invisible . Our body of work combines expert advancement on an overt problem in the field with aesthetic creative visualization . ”
Moving images of superhero Ironman , basketball game superstar Steph Curry , and even the abstractionist central characters in famous Picasso and Matisse artworks highlight the algorithm ’s ability to bring still images to life . It works by first discover the soul or object in an paradigm by sketch their body to make what researcher call a “ mask ” . A 3D template is then equal to and wrapped around the body position so that the target contain on a multi - dimensional role over a two - dimensional background . Information is store in each pixel to trail the distance between different points on the object ’s body for unstable motion that is maintained when the algorithm then pose the original texture ( people of color and patterns ) back over the under wrap .
It was once thought insufferable , but the research worker write in their study , available to show on pre - print sitearXiv , that their work can help shaver bring their artwork to life history , add depth to museum exhibits , and even provide realistic approaches to video game experiences .
“ There is some previous piece of work that tries to create a 3D character using multiple stand , ” say co - authorBrian Curless . “ But you still could n’t bring in someone to life and have them prevail out of a scene , and you could n’t bring AR into it . It was really surprising that we could get some compelling results with using just one photo . ”
Photo Wake - Up currently works good with onward - facing images . The researchers will be present their workplace at the annualConference on Computer Vision and Pattern Recognitionin southerly California later this calendar month .
