Last updated 08-15-2026
Category:
Reviews:
Join thousands of AI enthusiasts in the World of AI!
Drag Your GAN
Drag Your GAN lets you reshape a generated image by dragging handle points to target spots on the canvas. Instead of training on labeled pose data or 3D priors, DragGAN moves pixels inside a pretrained GAN's image manifold so edits stay photorealistic even when occluded regions get hallucinated. The project page hosts the SIGGRAPH 2023 paper, demo clips, and links to open source code.
Most GAN editors rely on fixed training annotations or 3D meshes, which limits what you can deform. DragGAN pairs feature-based motion supervision with discriminative point tracking so handle points follow your cursor while staying on the object surface. That makes it useful for nudging animal poses, facial expressions, car angles, and landscape layouts without retraining the generator.
Researchers, graphics students, and computer vision engineers use it to prototype interactive image editing workflows. The Max Planck Institute page distributes the paper PDF and GitHub repo under a non-commercial Creative Commons license, with contact details for the lead author.
Drag handle points to target positions for precise pixel-level control
Feature-based motion supervision drives handles without 3D priors
Point tracking uses discriminative GAN features to localize handles
Deforms animals, cars, humans, landscapes, and inverted real photos
Open source code published at github.com/XingangPan/DragGAN
Paper and demo videos from SIGGRAPH 2023 Conference Proceedings
Precise interactive control without manual training annotations
Produces realistic edits inside the GAN image manifold
Handles occluded content and rigid shape deformations
Open source code and paper available for free
Works across animals, humans, vehicles, and landscapes
Non-commercial license limits commercial deployment
Requires GAN expertise to run the research codebase
No hosted web app for casual users on the project page
Research project page with last update in May 2023
Is Drag Your GAN free to use?
Yes, Drag Your GAN is free for non-commercial use. The project page distributes the paper PDF and GitHub code under the Creative Commons CC BY-NC 4.0 license, with attribution required to the paper title and authors.
What does Drag Your GAN do?
Drag Your GAN lets you drag points on a generated image so pixels move to user-defined targets. DragGAN combines motion supervision and point tracking inside a GAN latent space to deform pose, shape, expression, and layout.
Where can I download Drag Your GAN code?
Drag Your GAN links to the open source repository at github.com/XingangPan/DragGAN from its project page. The same page also hosts the SIGGRAPH 2023 paper PDF for download.
What image types does Drag Your GAN support?
Drag Your GAN demo clips cover lions, cats, dogs, horses, elephants, human faces, cars, microscopes, and landscapes. The method also shows manipulation of real photos through GAN inversion.
Who created Drag Your GAN?
Drag Your GAN was authored by researchers from the Max Planck Institute for Informatics, MIT, University of Pennsylvania, and Google AR/VR. Xingang Pan is listed as the contact on the project page.
Can Drag Your GAN be used commercially?
Drag Your GAN site materials are released under CC BY-NC 4.0, which restricts commercial use. Review the license file on the project page before using images, text, or video in commercial projects.