Depth2Image

Instead of simply adding noise to the original image and pushing the new image towards the desired outcome, Depth2Image first estimates the depth of the picture to create a “depth-mask”. This depth-mask is then used as a mask to ensure the structure of the image remains intact. With Depth2Image, a variety of creative possibilities can be explored, providing transformation results that differ greatly from the original image but still keep the depth and coherence of the image. Sources for further information can be found below: