cuuupid / idm-vton

Best-in-class clothing virtual try on in the wild (non-commercial use only)

Warm

Public
660.8K runs
A100 (80GB)
GitHub
Paper
License

Run with an API

Playground API Examples README Versions

Input

garm_img

*file

Preview

Garment, should match the category, can be a product image or even a photo of someone

garment_des

string

Shift + Return to add a new line

Description of garment e.g. Short Sleeve Round Neck T-shirt

human_img

*file

Preview

Model, if this is not 3:4 check crop

category

string

Category of garment

Default: "upper_body"

steps

integer

(minimum: 1, maximum: 40)

Default: 30

Including mask_img and 4 more...

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

npx create-replicate --model=cuuupid/idm-vton

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 cuuupid/idm-vton using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.

const output = await replicate.run(
  "cuuupid/idm-vton:0513734a452173b8173e907e3a59d19a36266e55b48528559432bd21c7d7e985",
  {
    input: {
      crop: false,
      seed: 42,
      steps: 30,
      category: "upper_body",
      force_dc: false,
      garm_img: "https://replicate.delivery/pbxt/KgwTlZyFx5aUU3gc5gMiKuD5nNPTgliMlLUWx160G4z99YjO/sweater.webp",
      human_img: "https://replicate.delivery/pbxt/KgwTlhCMvDagRrcVzZJbuozNJ8esPqiNAIJS3eMgHrYuHmW4/KakaoTalk_Photo_2024-04-04-21-44-45.png",
      mask_only: false,
      garment_des: "cute pink top"
    }
  }
);
console.log(output);

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 cuuupid/idm-vton using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.

output = replicate.run(
    "cuuupid/idm-vton:0513734a452173b8173e907e3a59d19a36266e55b48528559432bd21c7d7e985",
    input={
        "crop": False,
        "seed": 42,
        "steps": 30,
        "category": "upper_body",
        "force_dc": False,
        "garm_img": "https://replicate.delivery/pbxt/KgwTlZyFx5aUU3gc5gMiKuD5nNPTgliMlLUWx160G4z99YjO/sweater.webp",
        "human_img": "https://replicate.delivery/pbxt/KgwTlhCMvDagRrcVzZJbuozNJ8esPqiNAIJS3eMgHrYuHmW4/KakaoTalk_Photo_2024-04-04-21-44-45.png",
        "mask_only": False,
        "garment_des": "cute pink top"
    }
)
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 cuuupid/idm-vton 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": "0513734a452173b8173e907e3a59d19a36266e55b48528559432bd21c7d7e985",
    "input": {
      "crop": false,
      "seed": 42,
      "steps": 30,
      "category": "upper_body",
      "force_dc": false,
      "garm_img": "https://replicate.delivery/pbxt/KgwTlZyFx5aUU3gc5gMiKuD5nNPTgliMlLUWx160G4z99YjO/sweater.webp",
      "human_img": "https://replicate.delivery/pbxt/KgwTlhCMvDagRrcVzZJbuozNJ8esPqiNAIJS3eMgHrYuHmW4/KakaoTalk_Photo_2024-04-04-21-44-45.png",
      "mask_only": false,
      "garment_des": "cute pink top"
    }
  }' \
  https://api.replicate.com/v1/predictions

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

You can run this model locally using Cog. First, install Cog:

brew install cog

If you don’t have Homebrew, there are other installation options available.

Run this to download the model and run it in your local environment:

cog predict r8.im/cuuupid/idm-vton@sha256:0513734a452173b8173e907e3a59d19a36266e55b48528559432bd21c7d7e985 \
  -i 'crop=false' \
  -i 'seed=42' \
  -i 'steps=30' \
  -i 'category="upper_body"' \
  -i 'force_dc=false' \
  -i 'garm_img="https://replicate.delivery/pbxt/KgwTlZyFx5aUU3gc5gMiKuD5nNPTgliMlLUWx160G4z99YjO/sweater.webp"' \
  -i 'human_img="https://replicate.delivery/pbxt/KgwTlhCMvDagRrcVzZJbuozNJ8esPqiNAIJS3eMgHrYuHmW4/KakaoTalk_Photo_2024-04-04-21-44-45.png"' \
  -i 'mask_only=false' \
  -i 'garment_des="cute pink top"'

To learn more, take a look at the Cog documentation.

Run this to download the model and run it in your local environment:

docker run -d -p 5000:5000 --gpus=all r8.im/cuuupid/idm-vton@sha256:0513734a452173b8173e907e3a59d19a36266e55b48528559432bd21c7d7e985
curl -s -X POST \
  -H "Content-Type: application/json" \
  -d $'{
    "input": {
      "crop": false,
      "seed": 42,
      "steps": 30,
      "category": "upper_body",
      "force_dc": false,
      "garm_img": "https://replicate.delivery/pbxt/KgwTlZyFx5aUU3gc5gMiKuD5nNPTgliMlLUWx160G4z99YjO/sweater.webp",
      "human_img": "https://replicate.delivery/pbxt/KgwTlhCMvDagRrcVzZJbuozNJ8esPqiNAIJS3eMgHrYuHmW4/KakaoTalk_Photo_2024-04-04-21-44-45.png",
      "mask_only": false,
      "garment_des": "cute pink top"
    }
  }' \
  http://localhost:5000/predictions

To learn more, take a look at the Cog documentation.

Output

{
  "completed_at": "2024-04-05T01:45:23.625048Z",
  "created_at": "2024-04-05T01:45:07.025000Z",
  "data_removed": false,
  "error": null,
  "id": "chz7vk2mt5rgg0cenfqaj2war8",
  "input": {
    "seed": 42,
    "steps": 30,
    "garm_img": "https://replicate.delivery/pbxt/KgwTlZyFx5aUU3gc5gMiKuD5nNPTgliMlLUWx160G4z99YjO/sweater.webp",
    "human_img": "https://replicate.delivery/pbxt/KgwTlhCMvDagRrcVzZJbuozNJ8esPqiNAIJS3eMgHrYuHmW4/KakaoTalk_Photo_2024-04-04-21-44-45.png",
    "garment_des": "cute pink top"
  },
  "logs": "0%|          | 0/1 [00:00<?, ?it/s]\n100%|██████████| 1/1 [00:03<00:00,  3.51s/it]\n100%|██████████| 1/1 [00:03<00:00,  3.51s/it]\n  0%|          | 0/1 [00:00<?, ?it/s]\n100%|██████████| 1/1 [00:03<00:00,  3.50s/it]\n100%|██████████| 1/1 [00:03<00:00,  3.50s/it]\n  0%|          | 0/30 [00:00<?, ?it/s]\n  3%|▎         | 1/30 [00:00<00:06,  4.49it/s]\n  7%|▋         | 2/30 [00:00<00:06,  4.46it/s]\n 10%|█         | 3/30 [00:00<00:05,  4.50it/s]\n 13%|█▎        | 4/30 [00:00<00:05,  4.49it/s]\n 17%|█▋        | 5/30 [00:01<00:05,  4.51it/s]\n 20%|██        | 6/30 [00:01<00:05,  4.53it/s]\n 23%|██▎       | 7/30 [00:01<00:05,  4.53it/s]\n 27%|██▋       | 8/30 [00:01<00:04,  4.53it/s]\n 30%|███       | 9/30 [00:01<00:04,  4.52it/s]\n 33%|███▎      | 10/30 [00:02<00:04,  4.52it/s]\n 37%|███▋      | 11/30 [00:02<00:04,  4.51it/s]\n 40%|████      | 12/30 [00:02<00:04,  4.48it/s]\n 43%|████▎     | 13/30 [00:02<00:03,  4.48it/s]\n 47%|████▋     | 14/30 [00:03<00:03,  4.51it/s]\n 50%|█████     | 15/30 [00:03<00:03,  4.51it/s]\n 53%|█████▎    | 16/30 [00:03<00:03,  4.51it/s]\n 57%|█████▋    | 17/30 [00:03<00:02,  4.50it/s]\n 60%|██████    | 18/30 [00:03<00:02,  4.50it/s]\n 63%|██████▎   | 19/30 [00:04<00:02,  4.49it/s]\n 67%|██████▋   | 20/30 [00:04<00:02,  4.47it/s]\n 70%|███████   | 21/30 [00:04<00:02,  4.47it/s]\n 73%|███████▎  | 22/30 [00:04<00:01,  4.46it/s]\n 77%|███████▋  | 23/30 [00:05<00:01,  4.47it/s]\n 80%|████████  | 24/30 [00:05<00:01,  4.47it/s]\n 83%|████████▎ | 25/30 [00:05<00:01,  4.49it/s]\n 87%|████████▋ | 26/30 [00:05<00:00,  4.49it/s]\n 90%|█████████ | 27/30 [00:06<00:00,  4.49it/s]\n 93%|█████████▎| 28/30 [00:06<00:00,  4.50it/s]\n 97%|█████████▋| 29/30 [00:06<00:00,  4.50it/s]\n100%|██████████| 30/30 [00:06<00:00,  4.50it/s]\n100%|██████████| 30/30 [00:06<00:00,  4.50it/s]",
  "metrics": {
    "predict_time": 16.590835,
    "total_time": 16.600048
  },
  "output": "https://replicate.delivery/pbxt/Tfs5JETdzlURKyKeUOltKwch7udHf8e2xmNsftwid7ca2E6UC/output.jpg",
  "started_at": "2024-04-05T01:45:07.034213Z",
  "status": "succeeded",
  "urls": {
    "get": "https://api.replicate.com/v1/predictions/chz7vk2mt5rgg0cenfqaj2war8",
    "cancel": "https://api.replicate.com/v1/predictions/chz7vk2mt5rgg0cenfqaj2war8/cancel"
  },
  "version": "e3893af4fb4bd5741752b35b395348c5f7a9ab5c4776264f5d38e41418081ed7"
}

Generated in

16.6 seconds

Tweak it ShareReport

0%|          | 0/1 [00:00<?, ?it/s]
100%|██████████| 1/1 [00:03<00:00,  3.51s/it]
100%|██████████| 1/1 [00:03<00:00,  3.51s/it]
  0%|          | 0/1 [00:00<?, ?it/s]
100%|██████████| 1/1 [00:03<00:00,  3.50s/it]
100%|██████████| 1/1 [00:03<00:00,  3.50s/it]
  0%|          | 0/30 [00:00<?, ?it/s]
  3%|▎         | 1/30 [00:00<00:06,  4.49it/s]
  7%|▋         | 2/30 [00:00<00:06,  4.46it/s]
 10%|█         | 3/30 [00:00<00:05,  4.50it/s]
 13%|█▎        | 4/30 [00:00<00:05,  4.49it/s]
 17%|█▋        | 5/30 [00:01<00:05,  4.51it/s]
 20%|██        | 6/30 [00:01<00:05,  4.53it/s]
 23%|██▎       | 7/30 [00:01<00:05,  4.53it/s]
 27%|██▋       | 8/30 [00:01<00:04,  4.53it/s]
 30%|███       | 9/30 [00:01<00:04,  4.52it/s]
 33%|███▎      | 10/30 [00:02<00:04,  4.52it/s]
 37%|███▋      | 11/30 [00:02<00:04,  4.51it/s]
 40%|████      | 12/30 [00:02<00:04,  4.48it/s]
 43%|████▎     | 13/30 [00:02<00:03,  4.48it/s]
 47%|████▋     | 14/30 [00:03<00:03,  4.51it/s]
 50%|█████     | 15/30 [00:03<00:03,  4.51it/s]
 53%|█████▎    | 16/30 [00:03<00:03,  4.51it/s]
 57%|█████▋    | 17/30 [00:03<00:02,  4.50it/s]
 60%|██████    | 18/30 [00:03<00:02,  4.50it/s]
 63%|██████▎   | 19/30 [00:04<00:02,  4.49it/s]
 67%|██████▋   | 20/30 [00:04<00:02,  4.47it/s]
 70%|███████   | 21/30 [00:04<00:02,  4.47it/s]
 73%|███████▎  | 22/30 [00:04<00:01,  4.46it/s]
 77%|███████▋  | 23/30 [00:05<00:01,  4.47it/s]
 80%|████████  | 24/30 [00:05<00:01,  4.47it/s]
 83%|████████▎ | 25/30 [00:05<00:01,  4.49it/s]
 87%|████████▋ | 26/30 [00:05<00:00,  4.49it/s]
 90%|█████████ | 27/30 [00:06<00:00,  4.49it/s]
 93%|█████████▎| 28/30 [00:06<00:00,  4.50it/s]
 97%|█████████▋| 29/30 [00:06<00:00,  4.50it/s]
100%|██████████| 30/30 [00:06<00:00,  4.50it/s]
100%|██████████| 30/30 [00:06<00:00,  4.50it/s]

This example was created by a different version, cuuupid/idm-vton:e3893af4.

Examples

View more examples

Run time and cost

This model costs approximately $0.024 to run on Replicate, or 41 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 A100 (80GB) GPU hardware. Predictions typically complete within 18 seconds.

Readme

Non-Commercial use only!

This is the current best-in-class virtual try-on model, created by the Korea Advanced Institute of Science & Technology (KAIST). It’s capable of virtual try-on “in the wild” which has notoriously been difficult for generative models to tackle, until now!

IDM-VTON : Improving Diffusion Models for Authentic Virtual Try-on in the Wild

This is an official implementation of paper ‘Improving Diffusion Models for Authentic Virtual Try-on in the Wild’ - paper - project page

teaser teaser2

TODO LIST

[x] demo model
[x] inference code
[ ] training code

Acknowledgements

For the demo, auto masking generation codes are based on OOTDiffusion and DCI-VTON.
Parts of the code are based on IP-Adapter.

Citation

@article{choi2024improving,
  title={Improving Diffusion Models for Virtual Try-on},
  author={Choi, Yisol and Kwak, Sangkyung and Lee, Kyungmin and Choi, Hyungwon and Shin, Jinwoo},
  journal={arXiv preprint arXiv:2403.05139},
  year={2024}
}

License

The codes and checkpoints in this repository are under the CC BY-NC-SA 4.0 license.