You're looking at a specific version of this model. Jump to the model overview.
Input schema
The fields you can use to run this model with an API. If you don’t give a value for a field its default value will be used.
| Field | Type | Default value | Description |
|---|---|---|---|
| image |
string
|
Input image. Facial images are expected to be aligned. If not, you can use https://replicate.com/cjwbw/face-align-cog to align your image first.
|
|
| model |
string
(enum)
|
CelebA-HQ
Options: CelebA-HQ, ImageNet, Places |
Choose a model depending on the input image.
|
| mask |
string
(enum)
|
face mask
Options: face mask, thick, thin, every second line, super-resolution, expand, half |
Choose a type for masking the image before repainting. Please refer to the Examples to see what each type of the mask looks like.
|
| steps |
integer
|
200
Min: 50 Max: 250 |
Reduce the total number of steps (without resampling) for faster inference. The lower it is, the more noise gets removed per step.
|
Output schema
The shape of the response you’ll get when you run this model with an API.
Schema
{'properties': {'inpaint': {'format': 'uri',
'title': 'Inpaint',
'type': 'string'},
'mask': {'format': 'uri', 'title': 'Mask', 'type': 'string'},
'masked_image': {'format': 'uri',
'title': 'Masked Image',
'type': 'string'}},
'required': ['mask', 'masked_image', 'inpaint'],
'title': 'Output',
'type': 'object'}