With just a single image as input, enables you to effortlessly obtain multiple angled views and generate intricate 3D videos.
SV3D surpasses existing open-source alternatives in generating detailed and faithful multi-views.
Offers superior pose-controllability and consistent object appearance across multiple views.
Produces accurate and realistic 3D meshes directly from single image inputs.
Can be used for commercial purposes with a Stability AI Membership.
Model weights available on Hugging Face for non-commercial use.
SV3D can create realistic and consistent views of an object from any angle, even those not present in the original image.
Leverages multi-view consistency to generate high-quality 3D meshes directly from novel views.
Enhances 3D quality in regions not visible in the predicted views.
Reduces baked-in lighting issues, leading to more realistic 3D models.
SV3D_u: Generates orbital videos from single images without camera conditioning. SV3D_p: Creates 3D video along specified camera paths using either single images or orbital views.
Variant | Input | Output | Camera path control |
---|---|---|---|
SV3D_u | Single image | Orbital video | No |
SV3D_p | Single image or orbital views | 3D video | Yes |