An imaginative neural network created an image that could be a blurry dashcam photo, or the work of one of those apps that turns snaps into paintings by stitching together its memories of real streets it was trained on. Nothing in the image actually exists, says Qifeng Chen at Stanford University, California, and Intel. Instead, his AI works from rough layouts that tell it what should be in each part of an image. Chen says the technique could eventually create game worlds that look truly real.
© 2001-2024 Fundación Dialnet · Todos los derechos reservados