okaris/omni-zero-couples 🔢📝🖼️ → 🖼️

▶️ 32.5K runs 📅 Sep 2024 ⚙️ Cog 0.9.24 🔗 GitHub ⚖️ License
image-consistent-character-generation image-style-transfer image-to-image

About

Omni-Zero Couples: A diffusion pipeline for zero-shot stylized couples portrait creation.

Example Output

Prompt:

"Cinematic still photo of a couple. emotional, harmonious, vignette, 4k epic detailed, shot on kodak, 35mm photo, sharp focus, high budget, cinemascope, moody, epic, gorgeous, film grain, grainy"

Output

Example output

Performance Metrics

12.61s Prediction Time
266.23s Total Time
All Input Parameters
{
  "seed": -1,
  "prompt": "Cinematic still photo of a couple. emotional, harmonious, vignette, 4k epic detailed, shot on kodak, 35mm photo, sharp focus, high budget, cinemascope, moody, epic, gorgeous, film grain, grainy",
  "base_image": "https://cdn-prod.styleof.com/inferences/cm1ho5cjl14nh14jec6phg2h8/i6k59e7gpsr45ufc7l8kun0g-medium.jpeg",
  "style_image": "https://cdn-prod.styleof.com/inferences/cm1ho5cjl14nh14jec6phg2h8/i6k59e7gpsr45ufc7l8kun0g-medium.jpeg",
  "guidance_scale": 3,
  "negative_prompt": "anime, cartoon, graphic, (blur, blurry, bokeh), text, painting, crayon, graphite, abstract, glitch, deformed, mutated, ugly, disfigured",
  "number_of_steps": 10,
  "identity_image_1": "https://cdn-prod.styleof.com/inferences/cm1hp4lea14oz14jeoghnex7g/dlgc5xwo0qzey7qaixy45i1o-medium.jpeg",
  "identity_image_2": "https://cdn-prod.styleof.com/inferences/cm1ho69ha14np14jesnusqiep/mp3aaktzqz20ujco5i3bi5s1-medium.jpeg",
  "number_of_images": 1,
  "mask_guidance_end": 1,
  "base_image_strength": 0.3,
  "mask_guidance_start": 0,
  "depth_image_strength": 0.2,
  "style_image_strength": 1,
  "identity_image_strength_1": 1,
  "identity_image_strength_2": 1
}
Input Parameters
seed Type: integerDefault: -1
Random seed for the model. Use -1 for random
prompt Type: stringDefault: Cinematic still photo of a couple. emotional, harmonious, vignette, 4k epic detailed, shot on kodak, 35mm photo, sharp focus, high budget, cinemascope, moody, epic, gorgeous, film grain, grainy
Prompt for the model
base_image Type: string
Base image for the model
depth_image Type: string
Depth image for the model
style_image Type: string
Style image for the model
guidance_scale Type: numberDefault: 3Range: 0 - 14
Guidance scale for the model
negative_prompt Type: stringDefault: anime, cartoon, graphic, (blur, blurry, bokeh), text, painting, crayon, graphite, abstract, glitch, deformed, mutated, ugly, disfigured
Negative prompt for the model
number_of_steps Type: integerDefault: 10Range: 1 - 50
Number of steps for the model
identity_image_1 Type: string
First identity image for the model
identity_image_2 Type: string
Second identity image for the model
number_of_images Type: integerDefault: 1Range: 1 - 4
Number of images to generate
mask_guidance_end Type: numberDefault: 1Range: 0 - 1
Mask guidance end value
base_image_strength Type: numberDefault: 0.2Range: 0 - 1
Base image strength for the model
mask_guidance_start Type: numberDefault: 0Range: 0 - 1
Mask guidance start value
depth_image_strength Type: numberDefault: 0.2Range: 0 - 1
Depth image strength for the model
style_image_strength Type: numberDefault: 1Range: 0 - 1
Style image strength for the model
identity_image_strength_1 Type: numberDefault: 1Range: 0 - 1
First identity image strength for the model
identity_image_strength_2 Type: numberDefault: 1Range: 0 - 1
Second identity image strength for the model
Output Schema

Output

Type: arrayItems Type: stringItems Format: uri

Example Execution Logs
base_image <PIL.JpegImagePlugin.JpegImageFile image mode=RGB size=960x720 at 0x711A70743410>
style_image <PIL.JpegImagePlugin.JpegImageFile image mode=RGB size=960x720 at 0x711B65DC6810>
identity_image_1 <PIL.JpegImagePlugin.JpegImageFile image mode=RGB size=960x720 at 0x711B65DC49D0>
identity_image_2 <PIL.JpegImagePlugin.JpegImageFile image mode=RGB size=960x720 at 0x71197C256B90>
depth_image None
/root/.pyenv/versions/3.11.10/lib/python3.11/site-packages/insightface/utils/transform.py:68: FutureWarning: `rcond` parameter will change to the default of machine precision times ``max(M, N)`` where M and N are the input matrix dimensions.
To use the future default and silence this warning we advise to pass `rcond=None`, to keep using the old, explicitly pass `rcond=-1`.
P = np.linalg.lstsq(X_homo, Y)[0].T # Affine matrix. 3 x 4
Face 1 -0: Age: 28, Gender: 0
Face 2 -0: Age: 76, Gender: 1
  0%|          | 0/10 [00:00<?, ?it/s]/root/.pyenv/versions/3.11.10/lib/python3.11/site-packages/torchsde/_brownian/brownian_interval.py:608: UserWarning: Should have tb<=t1 but got tb=2.3583380698653342 and t1=2.358338.
warnings.warn(f"Should have {tb_name}<=t1 but got {tb_name}={tb} and t1={self._end}.")
 10%|█         | 1/10 [00:00<00:05,  1.74it/s]
 20%|██        | 2/10 [00:01<00:04,  1.87it/s]
 30%|███       | 3/10 [00:01<00:03,  1.93it/s]
 40%|████      | 4/10 [00:02<00:03,  1.92it/s]
 50%|█████     | 5/10 [00:02<00:02,  1.95it/s]
 60%|██████    | 6/10 [00:03<00:02,  1.87it/s]
 70%|███████   | 7/10 [00:03<00:01,  1.91it/s]
 80%|████████  | 8/10 [00:04<00:01,  1.91it/s]
 90%|█████████ | 9/10 [00:04<00:00,  1.94it/s]/root/.pyenv/versions/3.11.10/lib/python3.11/site-packages/torchsde/_brownian/brownian_interval.py:599: UserWarning: Should have ta>=t0 but got ta=0.0 and t0=0.075972.
warnings.warn(f"Should have ta>=t0 but got ta={ta} and t0={self._start}.")
100%|██████████| 10/10 [00:05<00:00,  1.94it/s]
100%|██████████| 10/10 [00:05<00:00,  1.92it/s]
Version Details
Version ID
24ed4a267cd458569aa9f4da971ccc8d3587eaa47c43cf79572c3c7738a390f9
Version Created
September 25, 2024
Run on Replicate →