File size: 1,371 Bytes
c671618
 
6603155
 
9323e29
2ad7912
 
 
 
f18473a
 
2ad7912
 
f18473a
2ad7912
f18473a
 
 
2ad7912
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
f18473a
2ad7912
 
f18473a
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
---
license: mit
tags:
- object insertion
pipeline_tag: image-to-image
---

These are the model weights for [EraseDraw](https://erasedraw.cs.columbia.edu). 

<img src="https://erasedraw.cs.columbia.edu/static/img/samples_overview.png" alt="Sample Overview">

To use this model, install diffusers using main for now. The API is the same as that of InstructPix2Pix

```
pip install diffusers accelerate safetensors transformers
```

```
import PIL
import requests
import torch
from diffusers import StableDiffusionInstructPix2PixPipeline, EulerAncestralDiscreteScheduler

model_id = "alpercanberk/erasedraw"
pipe = StableDiffusionInstructPix2PixPipeline.from_pretrained(model_id, torch_dtype=torch.float16, safety_checker=None)
pipe.to("cuda")
pipe.scheduler = EulerAncestralDiscreteScheduler.from_config(pipe.scheduler.config)

url = "https://raw.githubusercontent.com/timothybrooks/instruct-pix2pix/main/imgs/example.jpg"
def download_image(url):
    image = PIL.Image.open(requests.get(url, stream=True).raw)
    image = PIL.ImageOps.exif_transpose(image)
    image = image.convert("RGB")
    return image
image = download_image(url)

prompt = "add sunglasses"
images = pipe(prompt, image=image, num_inference_steps=10, image_guidance_scale=1).images
images[0]
```


Code and data are coming soon to GitHub (find link on [website](https://erasedraw.cs.columbia.edu)) .