AnyDoor is a diffusion-based image generator with the power to teleport target objects to new scenes at user-specified locations in a harmonious way. Instead of tuning parameters for each object, our model is trained only once and effortlessly generalizes to diverse object-scene combinations at the inference stage. Such a challenging zero-shot setting requires an adequate characterization of a certain object. To this end, we complement the commonly used identity feature with detail features, which are carefully designed to maintain texture details yet allow versatile local variations (e.g., lighting, orientation, posture, etc.), supporting the object in favorably blending with different surroundings.


Yes, the website is free to use.
Sometimes there could be server error and you can try to chat again or later. If you are seeing the error after you are using the tool for a bit, the cause could be that we have set a limit of requests per day for each user to ensure fair access to everyone. We plan to expand these limits soon.
You can contact us for any questions or suggestions at [email protected].