lucataco/ip_adapter-face-inpaint – Run with an API on Replicate

lucataco / ip_adapter-face-inpaint

A combination of ip_adapter SDv1.5 and mediapipe-face to inpaint a face

Cold

Public
4K runs
L40S
GitHub
Paper
License

Run with an API

Playground API Examples README Versions

Input

Run this model in Node.js with one line of code:

npx create-replicate --model=lucataco/ip_adapter-face-inpaint

or set up a project from scratch

Install Replicate’s Node.js client library:

npm install replicate

Set the REPLICATE_API_TOKEN environment variable:

export REPLICATE_API_TOKEN=<paste-your-token-here>

Find your API token in your account settings.

Import and set up the client:

import Replicate from "replicate";

const replicate = new Replicate({
  auth: process.env.REPLICATE_API_TOKEN,
});

Run lucataco/ip_adapter-face-inpaint using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.

const output = await replicate.run(
  "lucataco/ip_adapter-face-inpaint:b199f118e2133894551cc59ff0777276e275cf64e9e8e0369ca6c4c599097890",
  {
    input: {
      prompt: "",
      strength: 0.7,
      face_image: "https://replicate.delivery/pbxt/JsWcNDv5Lp8NeokDpKZUYxdKLCRKzjr0c1hPuDfotlr4zLky/girl.png",
      blur_amount: 0,
      num_outputs: 1,
      source_image: "https://replicate.delivery/pbxt/JsWcMzZ0IfxfBPZBtgqtdTNsCYDFMeGxjy0kQld7cgZkuhvG/geisha.png"
    }
  }
);

// To access the file URL:
console.log(output[0].url()); //=> "http://example.com"

// To write the file to disk:
fs.writeFile("my-image.png", output[0]);

To learn more, take a look at the guide on getting started with Node.js.

Install Replicate’s Python client library:

pip install replicate

Set the REPLICATE_API_TOKEN environment variable:

export REPLICATE_API_TOKEN=<paste-your-token-here>

Find your API token in your account settings.

Import the client:

import replicate

Run lucataco/ip_adapter-face-inpaint using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.

output = replicate.run(
    "lucataco/ip_adapter-face-inpaint:b199f118e2133894551cc59ff0777276e275cf64e9e8e0369ca6c4c599097890",
    input={
        "prompt": "",
        "strength": 0.7,
        "face_image": "https://replicate.delivery/pbxt/JsWcNDv5Lp8NeokDpKZUYxdKLCRKzjr0c1hPuDfotlr4zLky/girl.png",
        "blur_amount": 0,
        "num_outputs": 1,
        "source_image": "https://replicate.delivery/pbxt/JsWcMzZ0IfxfBPZBtgqtdTNsCYDFMeGxjy0kQld7cgZkuhvG/geisha.png"
    }
)
print(output)

To learn more, take a look at the guide on getting started with Python.

Set the REPLICATE_API_TOKEN environment variable:

export REPLICATE_API_TOKEN=<paste-your-token-here>

Find your API token in your account settings.

Run lucataco/ip_adapter-face-inpaint using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.

curl -s -X POST \
  -H "Authorization: Bearer $REPLICATE_API_TOKEN" \
  -H "Content-Type: application/json" \
  -H "Prefer: wait" \
  -d $'{
    "version": "b199f118e2133894551cc59ff0777276e275cf64e9e8e0369ca6c4c599097890",
    "input": {
      "prompt": "",
      "strength": 0.7,
      "face_image": "https://replicate.delivery/pbxt/JsWcNDv5Lp8NeokDpKZUYxdKLCRKzjr0c1hPuDfotlr4zLky/girl.png",
      "blur_amount": 0,
      "num_outputs": 1,
      "source_image": "https://replicate.delivery/pbxt/JsWcMzZ0IfxfBPZBtgqtdTNsCYDFMeGxjy0kQld7cgZkuhvG/geisha.png"
    }
  }' \
  https://api.replicate.com/v1/predictions

To learn more, take a look at Replicate’s HTTP API reference docs.

Output

{
  "completed_at": "2023-11-15T00:51:17.675723Z",
  "created_at": "2023-11-15T00:51:04.601746Z",
  "data_removed": false,
  "error": null,
  "id": "4olm5h3bk6mnf7gimcd7uju5lq",
  "input": {
    "prompt": "",
    "strength": 0.7,
    "face_image": "https://replicate.delivery/pbxt/JsWcNDv5Lp8NeokDpKZUYxdKLCRKzjr0c1hPuDfotlr4zLky/girl.png",
    "blur_amount": 0,
    "num_outputs": 1,
    "source_image": "https://replicate.delivery/pbxt/JsWcMzZ0IfxfBPZBtgqtdTNsCYDFMeGxjy0kQld7cgZkuhvG/geisha.png"
  },
  "logs": "Using seed: 12755\n(122, 322, 382, 382)\n  0%|          | 0/35 [00:00<?, ?it/s]\n  3%|▎         | 1/35 [00:00<00:06,  5.42it/s]\n  6%|▌         | 2/35 [00:00<00:06,  5.39it/s]\n  9%|▊         | 3/35 [00:00<00:05,  5.37it/s]\n 11%|█▏        | 4/35 [00:00<00:05,  5.37it/s]\n 14%|█▍        | 5/35 [00:00<00:05,  5.35it/s]\n 17%|█▋        | 6/35 [00:01<00:05,  5.35it/s]\n 20%|██        | 7/35 [00:01<00:05,  5.36it/s]\n 23%|██▎       | 8/35 [00:01<00:05,  5.35it/s]\n 26%|██▌       | 9/35 [00:01<00:04,  5.35it/s]\n 29%|██▊       | 10/35 [00:01<00:04,  5.34it/s]\n 31%|███▏      | 11/35 [00:02<00:04,  5.33it/s]\n 34%|███▍      | 12/35 [00:02<00:04,  5.33it/s]\n 37%|███▋      | 13/35 [00:02<00:04,  5.34it/s]\n 40%|████      | 14/35 [00:02<00:03,  5.33it/s]\n 43%|████▎     | 15/35 [00:02<00:03,  5.34it/s]\n 46%|████▌     | 16/35 [00:02<00:03,  5.32it/s]\n 49%|████▊     | 17/35 [00:03<00:03,  5.32it/s]\n 51%|█████▏    | 18/35 [00:03<00:03,  5.32it/s]\n 54%|█████▍    | 19/35 [00:03<00:03,  5.32it/s]\n 57%|█████▋    | 20/35 [00:03<00:02,  5.33it/s]\n 60%|██████    | 21/35 [00:03<00:02,  5.32it/s]\n 63%|██████▎   | 22/35 [00:04<00:02,  5.32it/s]\n 66%|██████▌   | 23/35 [00:04<00:02,  5.32it/s]\n 69%|██████▊   | 24/35 [00:04<00:02,  5.32it/s]\n 71%|███████▏  | 25/35 [00:04<00:01,  5.33it/s]\n 74%|███████▍  | 26/35 [00:04<00:01,  5.32it/s]\n 77%|███████▋  | 27/35 [00:05<00:01,  5.31it/s]\n 80%|████████  | 28/35 [00:05<00:01,  5.32it/s]\n 83%|████████▎ | 29/35 [00:05<00:01,  5.31it/s]\n 86%|████████▌ | 30/35 [00:05<00:00,  5.32it/s]\n 89%|████████▊ | 31/35 [00:05<00:00,  5.31it/s]\n 91%|█████████▏| 32/35 [00:06<00:00,  5.31it/s]\n 94%|█████████▍| 33/35 [00:06<00:00,  5.31it/s]\n 97%|█████████▋| 34/35 [00:06<00:00,  5.31it/s]\n100%|██████████| 35/35 [00:06<00:00,  5.32it/s]\n100%|██████████| 35/35 [00:06<00:00,  5.33it/s]",
  "metrics": {
    "predict_time": 13.040806,
    "total_time": 13.073977
  },
  "output": [
    "https://replicate.delivery/pbxt/xEdYleGC352GAyx1ovitmxqg1Bh36JDESvKl3MkWgUFCQO8IA/out-0.png"
  ],
  "started_at": "2023-11-15T00:51:04.634917Z",
  "status": "succeeded",
  "urls": {
    "get": "https://api.replicate.com/v1/predictions/4olm5h3bk6mnf7gimcd7uju5lq",
    "cancel": "https://api.replicate.com/v1/predictions/4olm5h3bk6mnf7gimcd7uju5lq/cancel"
  },
  "version": "b199f118e2133894551cc59ff0777276e275cf64e9e8e0369ca6c4c599097890"
}

Generated in

13.1 seconds

Tweak itReport

Using seed: 12755
(122, 322, 382, 382)
  0%|          | 0/35 [00:00<?, ?it/s]
  3%|▎         | 1/35 [00:00<00:06,  5.42it/s]
  6%|▌         | 2/35 [00:00<00:06,  5.39it/s]
  9%|▊         | 3/35 [00:00<00:05,  5.37it/s]
 11%|█▏        | 4/35 [00:00<00:05,  5.37it/s]
 14%|█▍        | 5/35 [00:00<00:05,  5.35it/s]
 17%|█▋        | 6/35 [00:01<00:05,  5.35it/s]
 20%|██        | 7/35 [00:01<00:05,  5.36it/s]
 23%|██▎       | 8/35 [00:01<00:05,  5.35it/s]
 26%|██▌       | 9/35 [00:01<00:04,  5.35it/s]
 29%|██▊       | 10/35 [00:01<00:04,  5.34it/s]
 31%|███▏      | 11/35 [00:02<00:04,  5.33it/s]
 34%|███▍      | 12/35 [00:02<00:04,  5.33it/s]
 37%|███▋      | 13/35 [00:02<00:04,  5.34it/s]
 40%|████      | 14/35 [00:02<00:03,  5.33it/s]
 43%|████▎     | 15/35 [00:02<00:03,  5.34it/s]
 46%|████▌     | 16/35 [00:02<00:03,  5.32it/s]
 49%|████▊     | 17/35 [00:03<00:03,  5.32it/s]
 51%|█████▏    | 18/35 [00:03<00:03,  5.32it/s]
 54%|█████▍    | 19/35 [00:03<00:03,  5.32it/s]
 57%|█████▋    | 20/35 [00:03<00:02,  5.33it/s]
 60%|██████    | 21/35 [00:03<00:02,  5.32it/s]
 63%|██████▎   | 22/35 [00:04<00:02,  5.32it/s]
 66%|██████▌   | 23/35 [00:04<00:02,  5.32it/s]
 69%|██████▊   | 24/35 [00:04<00:02,  5.32it/s]
 71%|███████▏  | 25/35 [00:04<00:01,  5.33it/s]
 74%|███████▍  | 26/35 [00:04<00:01,  5.32it/s]
 77%|███████▋  | 27/35 [00:05<00:01,  5.31it/s]
 80%|████████  | 28/35 [00:05<00:01,  5.32it/s]
 83%|████████▎ | 29/35 [00:05<00:01,  5.31it/s]
 86%|████████▌ | 30/35 [00:05<00:00,  5.32it/s]
 89%|████████▊ | 31/35 [00:05<00:00,  5.31it/s]
 91%|█████████▏| 32/35 [00:06<00:00,  5.31it/s]
 94%|█████████▍| 33/35 [00:06<00:00,  5.31it/s]
 97%|█████████▋| 34/35 [00:06<00:00,  5.31it/s]
100%|██████████| 35/35 [00:06<00:00,  5.32it/s]
100%|██████████| 35/35 [00:06<00:00,  5.33it/s]

Examples

View more examples

Run time and cost

This model costs approximately $0.10 to run on Replicate, or 10 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 105 seconds. The predict time for this model varies significantly based on the inputs.

Readme

An attempt to combine the models: chigozienri/mediapipe-face and chigozienri/ip_adapter-face-inpaint to auto-inpaint a face from an input image to a target image

Note: I recommend to use 1024 SDXL image resolutions for input