fofr / expression-editor

Quickly edit the expression of a face

  • Public
  • 74.8K runs
  • GitHub
  • Weights
  • Paper
  • License
Iterate in playground

Input

image
file

Image of a face

number
(minimum: -20, maximum: 20)

Rotation pitch: Adjusts the up and down tilt of the face

Default: 0

number
(minimum: -20, maximum: 20)

Rotation yaw: Adjusts the left and right turn of the face

Default: 0

number
(minimum: -20, maximum: 20)

Rotation roll: Adjusts the tilt of the face to the left or right

Default: 0

number
(minimum: -20, maximum: 5)

Blink: Controls the degree of eye closure

Default: 0

number
(minimum: -10, maximum: 15)

Eyebrow: Adjusts the height and shape of the eyebrows

Default: 0

number
(minimum: 0, maximum: 25)

Wink: Controls the degree of one eye closing

Default: 0

number
(minimum: -15, maximum: 15)

Pupil X: Adjusts the horizontal position of the pupils

Default: 0

number
(minimum: -15, maximum: 15)

Pupil Y: Adjusts the vertical position of the pupils

Default: 0

number
(minimum: -30, maximum: 120)

AAA: Controls the mouth opening for 'aaa' sound

Default: 0

number
(minimum: -20, maximum: 15)

EEE: Controls the mouth shape for 'eee' sound

Default: 0

number
(minimum: -20, maximum: 15)

WOO: Controls the mouth shape for 'woo' sound

Default: 0

number
(minimum: -0.3, maximum: 1.3)

Smile: Adjusts the degree of smiling

Default: 0

number
(minimum: 0, maximum: 1)

Source ratio

Default: 1

number
(minimum: -0.2, maximum: 1.2)

Sample ratio

Default: 1

number
(minimum: 1.5, maximum: 2.5)

Crop factor

Default: 1.7

string

Format of the output images

Default: "webp"

integer
(minimum: 0, maximum: 100)

Quality of the output images, from 0 to 100. 100 is best quality, 0 is lowest quality.

Default: 95

Output

output
Generated in

This output was created using a different version of the model, fofr/expression-editor:949a601a.

Run time and cost

This model costs approximately $0.00098 to run on Replicate, or 1020 runs per $1, but this varies depending on your inputs. It is also open source and you can run it on your own computer with Docker.

This model runs on Nvidia L40S GPU hardware. Predictions typically complete within 1 seconds.