helios-infotech/sketch_to_image 🔢🖼️📝 → 🖼️

▶️ 2.5K runs 📅 Jun 2024 ⚙️ Cog 0.9.8

About

AI that transforms sketches into realistic images. Upload your drawing and describe it in the prompt. You can also adjust the ControlNet parameters and scale the image to a higher resolution for better results

Example Output

Prompt:

"a photo of interior of a home, sofa, carpet, realistic,4k"

Output

Performance Metrics

32.03s Prediction Time

190.16s Total Time

All Input Parameters

{
  "seed": 0,
  "image": "https://replicate.delivery/pbxt/L1czB5TWtOa0X7kdsFFhYpyX8yOR3k2VVh3vXnIdNkQylCu5/430d87ed19a5e7c4ef75158c271a45cb.jpg",
  "scale": 2,
  "prompt": "a photo of interior of a home, sofa, carpet, realistic,4k",
  "cn_depth_strength": 1,
  "cn_lineart_strength": 1,
  "num_inference_steps": 30
}

Input Parameters

seed Type: integerDefault: 0Range: 0 - ∞: Enter seed > 0,if you enter 0 it will generate random seed
image (required) Type: string: sketch image
scale Type: numberDefault: 1Range: 0 - 10: Factor to scale image by
prompt Type: stringDefault: realistic,4k
cn_depth_strength Type: numberDefault: 1Range: 0 - 1: Depth controlnet strength
cn_lineart_strength Type: numberDefault: 0.8Range: 0 - 1: Line art controlnet strength
num_inference_steps Type: integerDefault: 30Range: 0 - 60: Number of inference steps to generate image

Output Schema

Output

Type: string • Format: uri

Example Execution Logs

/root/.pyenv/versions/3.10.14/lib/python3.10/site-packages/torch/functional.py:512: UserWarning: torch.meshgrid: in an upcoming release, it will be required to pass the indexing argument. (Triggered internally at ../aten/src/ATen/native/TensorShape.cpp:3587.)
return _VF.meshgrid(tensors, **kwargs)  # type: ignore[attr-defined]
  0%|          | 0/30 [00:00<?, ?it/s]/root/.pyenv/versions/3.10.14/lib/python3.10/site-packages/torch/nn/modules/conv.py:456: UserWarning: Plan failed with a cudnnException: CUDNN_BACKEND_EXECUTION_PLAN_DESCRIPTOR: cudnnFinalize Descriptor Failed cudnn_status: CUDNN_STATUS_NOT_SUPPORTED (Triggered internally at ../aten/src/ATen/native/cudnn/Conv_v8.cpp:919.)
return F.conv2d(input, weight, bias, self.stride,
  3%|▎         | 1/30 [00:01<00:33,  1.16s/it]
 10%|█         | 3/30 [00:01<00:10,  2.64it/s]
 17%|█▋        | 5/30 [00:01<00:05,  4.31it/s]
 23%|██▎       | 7/30 [00:01<00:03,  5.80it/s]
 30%|███       | 9/30 [00:01<00:03,  7.00it/s]
 33%|███▎      | 10/30 [00:02<00:02,  7.44it/s]
 40%|████      | 12/30 [00:02<00:02,  8.40it/s]
 47%|████▋     | 14/30 [00:02<00:01,  8.93it/s]
 53%|█████▎    | 16/30 [00:02<00:01,  9.39it/s]
 60%|██████    | 18/30 [00:02<00:01,  9.78it/s]
 67%|██████▋   | 20/30 [00:02<00:00, 10.02it/s]
 73%|███████▎  | 22/30 [00:03<00:00, 10.19it/s]
 80%|████████  | 24/30 [00:03<00:00, 10.20it/s]
 87%|████████▋ | 26/30 [00:03<00:00, 10.32it/s]
 93%|█████████▎| 28/30 [00:03<00:00, 10.34it/s]
100%|██████████| 30/30 [00:03<00:00, 10.34it/s]
100%|██████████| 30/30 [00:03<00:00,  7.61it/s]

Version Details

Version ID: feb7325e48612a443356bff3d0e03af21a42570f87bee6e8ea4f275f2bd3e6f9
Version Created: September 1, 2024

Run on Replicate →