Generating 3D models from a 2D picture isn’t new by any means, but the technology behind the systems is indeed getting better. And as it does, the whole thing gets faster, and the produced materials get better. This new system in particular, designed by researchers from Waseda University in Japan, manages to create 3D head models in just a matter of seconds from a 2D image.
According to the system’s makers, the system is perfectly capable of converting the image in just 1.2 seconds. Furthermore, you don’t need any kind of special camera to make it happen. The researchers point out that you only need a standard webcam to make the magic happen. However, they also mention that taking an image with a better camera, like an SLR, the results do get better. But, as an added bonus, other equipment like range scanners aren’t needed to get the job done, and therefore should save costs a bit, as well as time.
They get the job done so quickly by utilizing a database of over 1,000 people. When the system analyzes a picture, it compares that image against the database of collected people, and starts comparing certain points on the scanned image. These characteristic elements are used to speed up the whole system, and therefore is able to almost instantly create the 3D model. As for usefulness, the designers say that this could be used for video game developers, especially those who would like to put the face of the gamer into the video game itself. Check out the video below to see it in action.