What is DragGAN?
DragGAN is a powerful tool for interactive point-based manipulation on the generative image manifold. It allows users to "drag" any points of the image to precisely reach target points in a user-interactive manner.
How does DragGAN work?
DragGAN consists of two main components: 1) a feature-based motion supervision that drives the handle point to move towards the target position, and 2) a new point tracking approach that leverages the discriminative GAN features to keep localizing the position of the handle points.
Features of DragGAN
- Precise control over where pixels go, thus manipulating the pose, shape, expression, and layout of diverse categories such as animals, cars, humans, landscapes, etc.
- Ability to deform an image with realistic outputs even for challenging scenarios such as hallucinating occluded content and deforming shapes that consistently follow the object's rigidity.
- Can be used for image manipulation and point tracking tasks.
Price
The paper and code are available for free, and the images, text, and video files on the site are made freely available for non-commercial use under the Creative Commons CC BY-NC 4.0 license.