DraGAN is an AI photo manipulation tool with the ability to control the pose, shape, expression, and layout of your images. Whether animals, cars, humans, landscapes, or the like, the tool can produce realistic outputs. You just need to “drag” the points on the image to precisely reach target points.
Its image manipulations are performed on the learned generative image manifold of a GAN. So the tool can take on challenging scenarios such as hallucinating occluded content. You can use its interactive manipulation to change how things look, feel, and where they are in the picture.
To use DraGAN, upload an image to it and place points on that image. After that, you can change the structure by dragging the points exactly where you want them to be. DragGAN will then automatically edit the image to match your changes.
Its other features include generating photorealistic images, 3D image transformation, and creative artistic editing.
Video demos of the tool are available on the project’s website, but it’s still under development (at the time of writing).