Spaces:
Paused
Paused
AndranikSargsyan
commited on
Commit
β’
f1cc496
1
Parent(s):
073105a
add support for diffusers checkpoint loading
Browse filesThis view is limited to 50 files because it contains too many changes. Β
See raw diff
- .gitignore +3 -1
- LICENSE +21 -0
- {assets β __assets__/demo}/config/ddpm/v1.yaml +0 -0
- {assets β __assets__/demo}/config/ddpm/v2-upsample.yaml +0 -0
- {assets β __assets__/demo}/config/encoders/clip.yaml +0 -0
- {assets β __assets__/demo}/config/encoders/openclip.yaml +0 -0
- {assets β __assets__/demo}/config/unet/inpainting/v1.yaml +0 -0
- {assets β __assets__/demo}/config/unet/inpainting/v2.yaml +0 -0
- {assets β __assets__/demo}/config/unet/upsample/v2.yaml +0 -0
- {assets β __assets__/demo}/config/vae-upsample.yaml +0 -0
- {assets β __assets__/demo}/config/vae.yaml +0 -0
- {assets β __assets__/demo}/examples/images_1024/a19.jpg +0 -0
- {assets β __assets__/demo}/examples/images_1024/a2.jpg +0 -0
- {assets β __assets__/demo}/examples/images_1024/a4.jpg +0 -0
- {assets β __assets__/demo}/examples/images_1024/a40.jpg +0 -0
- {assets β __assets__/demo}/examples/images_1024/a46.jpg +0 -0
- {assets β __assets__/demo}/examples/images_1024/a51.jpg +0 -0
- {assets β __assets__/demo}/examples/images_1024/a54.jpg +0 -0
- {assets β __assets__/demo}/examples/images_1024/a65.jpg +0 -0
- {assets β __assets__/demo}/examples/images_2048/a19.jpg +0 -0
- {assets β __assets__/demo}/examples/images_2048/a2.jpg +0 -0
- {assets β __assets__/demo}/examples/images_2048/a4.jpg +0 -0
- {assets β __assets__/demo}/examples/images_2048/a40.jpg +0 -0
- {assets β __assets__/demo}/examples/images_2048/a46.jpg +0 -0
- {assets β __assets__/demo}/examples/images_2048/a51.jpg +0 -0
- {assets β __assets__/demo}/examples/images_2048/a54.jpg +0 -0
- {assets β __assets__/demo}/examples/images_2048/a65.jpg +0 -0
- {assets β __assets__/demo}/examples/sbs/a19.png +0 -0
- {assets β __assets__/demo}/examples/sbs/a2.png +0 -0
- {assets β __assets__/demo}/examples/sbs/a4.png +0 -0
- {assets β __assets__/demo}/examples/sbs/a40.png +0 -0
- {assets β __assets__/demo}/examples/sbs/a46.png +0 -0
- {assets β __assets__/demo}/examples/sbs/a51.png +0 -0
- {assets β __assets__/demo}/examples/sbs/a54.png +0 -0
- {assets β __assets__/demo}/examples/sbs/a65.png +0 -0
- {assets β __assets__/demo}/sr_info.png +0 -0
- app.py +59 -98
- assets/.gitignore +0 -1
- config/ddpm/v1.yaml +14 -0
- config/ddpm/v2-upsample.yaml +24 -0
- config/encoders/clip.yaml +1 -0
- config/encoders/openclip.yaml +4 -0
- config/unet/inpainting/v1.yaml +15 -0
- config/unet/inpainting/v2.yaml +16 -0
- config/unet/upsample/v2.yaml +19 -0
- config/vae-upsample.yaml +16 -0
- config/vae.yaml +17 -0
- lib/models/__init__.py +0 -1
- lib/models/common.py +0 -49
- lib/models/ds_inp.py +0 -51
.gitignore
CHANGED
@@ -4,4 +4,6 @@
|
|
4 |
|
5 |
outputs/
|
6 |
gradio_tmp/
|
7 |
-
__pycache__/
|
|
|
|
|
|
4 |
|
5 |
outputs/
|
6 |
gradio_tmp/
|
7 |
+
__pycache__/
|
8 |
+
|
9 |
+
checkpoints/
|
LICENSE
ADDED
@@ -0,0 +1,21 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
MIT License
|
2 |
+
|
3 |
+
Copyright (c) 2023 Picsart AI Research (PAIR)
|
4 |
+
|
5 |
+
Permission is hereby granted, free of charge, to any person obtaining a copy
|
6 |
+
of this software and associated documentation files (the "Software"), to deal
|
7 |
+
in the Software without restriction, including without limitation the rights
|
8 |
+
to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
|
9 |
+
copies of the Software, and to permit persons to whom the Software is
|
10 |
+
furnished to do so, subject to the following conditions:
|
11 |
+
|
12 |
+
The above copyright notice and this permission notice shall be included in all
|
13 |
+
copies or substantial portions of the Software.
|
14 |
+
|
15 |
+
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
|
16 |
+
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
|
17 |
+
FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
|
18 |
+
AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
|
19 |
+
LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
|
20 |
+
OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
|
21 |
+
SOFTWARE.
|
{assets β __assets__/demo}/config/ddpm/v1.yaml
RENAMED
File without changes
|
{assets β __assets__/demo}/config/ddpm/v2-upsample.yaml
RENAMED
File without changes
|
{assets β __assets__/demo}/config/encoders/clip.yaml
RENAMED
File without changes
|
{assets β __assets__/demo}/config/encoders/openclip.yaml
RENAMED
File without changes
|
{assets β __assets__/demo}/config/unet/inpainting/v1.yaml
RENAMED
File without changes
|
{assets β __assets__/demo}/config/unet/inpainting/v2.yaml
RENAMED
File without changes
|
{assets β __assets__/demo}/config/unet/upsample/v2.yaml
RENAMED
File without changes
|
{assets β __assets__/demo}/config/vae-upsample.yaml
RENAMED
File without changes
|
{assets β __assets__/demo}/config/vae.yaml
RENAMED
File without changes
|
{assets β __assets__/demo}/examples/images_1024/a19.jpg
RENAMED
File without changes
|
{assets β __assets__/demo}/examples/images_1024/a2.jpg
RENAMED
File without changes
|
{assets β __assets__/demo}/examples/images_1024/a4.jpg
RENAMED
File without changes
|
{assets β __assets__/demo}/examples/images_1024/a40.jpg
RENAMED
File without changes
|
{assets β __assets__/demo}/examples/images_1024/a46.jpg
RENAMED
File without changes
|
{assets β __assets__/demo}/examples/images_1024/a51.jpg
RENAMED
File without changes
|
{assets β __assets__/demo}/examples/images_1024/a54.jpg
RENAMED
File without changes
|
{assets β __assets__/demo}/examples/images_1024/a65.jpg
RENAMED
File without changes
|
{assets β __assets__/demo}/examples/images_2048/a19.jpg
RENAMED
File without changes
|
{assets β __assets__/demo}/examples/images_2048/a2.jpg
RENAMED
File without changes
|
{assets β __assets__/demo}/examples/images_2048/a4.jpg
RENAMED
File without changes
|
{assets β __assets__/demo}/examples/images_2048/a40.jpg
RENAMED
File without changes
|
{assets β __assets__/demo}/examples/images_2048/a46.jpg
RENAMED
File without changes
|
{assets β __assets__/demo}/examples/images_2048/a51.jpg
RENAMED
File without changes
|
{assets β __assets__/demo}/examples/images_2048/a54.jpg
RENAMED
File without changes
|
{assets β __assets__/demo}/examples/images_2048/a65.jpg
RENAMED
File without changes
|
{assets β __assets__/demo}/examples/sbs/a19.png
RENAMED
File without changes
|
{assets β __assets__/demo}/examples/sbs/a2.png
RENAMED
File without changes
|
{assets β __assets__/demo}/examples/sbs/a4.png
RENAMED
File without changes
|
{assets β __assets__/demo}/examples/sbs/a40.png
RENAMED
File without changes
|
{assets β __assets__/demo}/examples/sbs/a46.png
RENAMED
File without changes
|
{assets β __assets__/demo}/examples/sbs/a51.png
RENAMED
File without changes
|
{assets β __assets__/demo}/examples/sbs/a54.png
RENAMED
File without changes
|
{assets β __assets__/demo}/examples/sbs/a65.png
RENAMED
File without changes
|
{assets β __assets__/demo}/sr_info.png
RENAMED
File without changes
|
app.py
CHANGED
@@ -1,40 +1,44 @@
|
|
1 |
import os
|
|
|
|
|
2 |
from collections import OrderedDict
|
3 |
|
4 |
import gradio as gr
|
5 |
import shutil
|
6 |
import uuid
|
7 |
import torch
|
8 |
-
from pathlib import Path
|
9 |
-
from lib.utils.iimage import IImage
|
10 |
from PIL import Image
|
11 |
|
12 |
-
|
13 |
-
|
14 |
-
|
|
|
|
|
|
|
15 |
|
16 |
|
17 |
-
TMP_DIR = 'gradio_tmp'
|
18 |
-
if
|
19 |
-
shutil.rmtree(TMP_DIR)
|
20 |
-
|
21 |
|
22 |
-
os.environ['GRADIO_TEMP_DIR'] = TMP_DIR
|
23 |
|
24 |
on_huggingspace = os.environ.get("SPACE_AUTHOR_NAME") == "PAIR"
|
25 |
|
26 |
negative_prompt_str = "text, bad anatomy, bad proportions, blurry, cropped, deformed, disfigured, duplicate, error, extra limbs, gross proportions, jpeg artifacts, long neck, low quality, lowres, malformed, morbid, mutated, mutilated, out of frame, ugly, worst quality"
|
27 |
positive_prompt_str = "Full HD, 4K, high quality, high resolution"
|
28 |
|
|
|
29 |
example_inputs = [
|
30 |
-
['
|
31 |
-
['
|
32 |
-
['
|
33 |
-
['
|
34 |
-
['
|
35 |
-
['
|
36 |
-
['
|
37 |
-
['
|
38 |
]
|
39 |
|
40 |
thumbnails = [
|
@@ -60,27 +64,35 @@ example_previews = [
|
|
60 |
]
|
61 |
|
62 |
# Load models
|
|
|
63 |
inpainting_models = OrderedDict([
|
64 |
-
("Dreamshaper Inpainting V8",
|
65 |
-
("Stable-Inpainting 2.0",
|
66 |
-
("Stable-Inpainting 1.5",
|
67 |
])
|
68 |
sr_model = models.sd2_sr.load_model(device='cuda:1')
|
69 |
sam_predictor = models.sam.load_model(device='cuda:0')
|
70 |
|
71 |
-
|
72 |
-
|
|
|
|
|
|
|
|
|
73 |
global inp_model
|
74 |
-
|
75 |
-
|
|
|
|
|
|
|
|
|
76 |
|
77 |
|
78 |
def save_user_session(hr_image, hr_mask, lr_results, prompt, session_id=None):
|
79 |
if session_id == '':
|
80 |
session_id = str(uuid.uuid4())
|
81 |
|
82 |
-
|
83 |
-
session_dir = tmp_dir / session_id
|
84 |
session_dir.mkdir(exist_ok=True, parents=True)
|
85 |
|
86 |
hr_image.save(session_dir / 'hr_image.png')
|
@@ -103,8 +115,7 @@ def recover_user_session(session_id):
|
|
103 |
if session_id == '':
|
104 |
return None, None, [], ''
|
105 |
|
106 |
-
|
107 |
-
session_dir = tmp_dir / session_id
|
108 |
lr_results_dir = session_dir / 'lr_results'
|
109 |
|
110 |
hr_image = Image.open(session_dir / 'hr_image.png')
|
@@ -121,64 +132,22 @@ def recover_user_session(session_id):
|
|
121 |
return hr_image, hr_mask, gallery, prompt
|
122 |
|
123 |
|
124 |
-
def
|
125 |
-
|
126 |
-
|
127 |
-
guidance_scale=7.5,
|
128 |
-
batch_size=1, session_id=''
|
129 |
):
|
130 |
torch.cuda.empty_cache()
|
|
|
131 |
|
132 |
-
|
133 |
-
batch_size = max(1, min(int(batch_size), 4))
|
134 |
-
|
135 |
-
image = IImage(hr_image).resize(512)
|
136 |
-
mask = IImage(imageMask['mask']).rgb().resize(512)
|
137 |
-
|
138 |
-
method = ['rasg']
|
139 |
if use_painta: method.append('painta')
|
|
|
140 |
method = '-'.join(method)
|
141 |
|
142 |
-
|
143 |
-
|
144 |
-
|
145 |
-
|
146 |
-
|
147 |
-
inpainted_image = rasg.run(
|
148 |
-
ddim=inp_model,
|
149 |
-
method=method,
|
150 |
-
prompt=prompt,
|
151 |
-
image=image,
|
152 |
-
mask=mask,
|
153 |
-
seed=seed,
|
154 |
-
eta=eta,
|
155 |
-
negative_prompt=negative_prompt,
|
156 |
-
positive_prompt=positive_prompt,
|
157 |
-
num_steps=ddim_steps,
|
158 |
-
guidance_scale=guidance_scale
|
159 |
-
).crop(image.size)
|
160 |
-
|
161 |
-
blended_image = poisson_blend(
|
162 |
-
orig_img=image.data[0],
|
163 |
-
fake_img=inpainted_image.data[0],
|
164 |
-
mask=mask.data[0],
|
165 |
-
dilation=12
|
166 |
-
)
|
167 |
-
blended_images.append(blended_image)
|
168 |
-
inpainted_images.append(inpainted_image.pil())
|
169 |
-
|
170 |
-
session_id = save_user_session(
|
171 |
-
hr_image, imageMask['mask'], inpainted_images, prompt, session_id=session_id)
|
172 |
-
|
173 |
-
return blended_images, session_id
|
174 |
-
|
175 |
-
|
176 |
-
def sd_run(use_painta, prompt, imageMask, hr_image, seed, eta,
|
177 |
-
negative_prompt, positive_prompt, ddim_steps,
|
178 |
-
guidance_scale=7.5,
|
179 |
-
batch_size=1, session_id=''
|
180 |
-
):
|
181 |
-
torch.cuda.empty_cache()
|
182 |
|
183 |
seed = int(seed)
|
184 |
batch_size = max(1, min(int(batch_size), 4))
|
@@ -195,7 +164,7 @@ def sd_run(use_painta, prompt, imageMask, hr_image, seed, eta,
|
|
195 |
for i in range(batch_size):
|
196 |
seed = seed + i * 1000
|
197 |
|
198 |
-
inpainted_image =
|
199 |
ddim=inp_model,
|
200 |
method=method,
|
201 |
prompt=prompt,
|
@@ -226,13 +195,12 @@ def sd_run(use_painta, prompt, imageMask, hr_image, seed, eta,
|
|
226 |
|
227 |
def upscale_run(
|
228 |
ddim_steps, seed, use_sam_mask, session_id, img_index,
|
229 |
-
negative_prompt='',
|
230 |
-
positive_prompt=', high resolution professional photo'
|
231 |
):
|
232 |
hr_image, hr_mask, gallery, prompt = recover_user_session(session_id)
|
233 |
|
234 |
if len(gallery) == 0:
|
235 |
-
return Image.open('
|
236 |
|
237 |
torch.cuda.empty_cache()
|
238 |
|
@@ -249,7 +217,7 @@ def upscale_run(
|
|
249 |
inpainted_image,
|
250 |
hr_image,
|
251 |
hr_mask,
|
252 |
-
prompt=prompt
|
253 |
noise_level=20,
|
254 |
blend_trick=True,
|
255 |
blend_output=True,
|
@@ -261,14 +229,7 @@ def upscale_run(
|
|
261 |
return output_image
|
262 |
|
263 |
|
264 |
-
|
265 |
-
set_model_from_name(model_name)
|
266 |
-
if use_rasg:
|
267 |
-
return rasg_run(*args)
|
268 |
-
return sd_run(*args)
|
269 |
-
|
270 |
-
|
271 |
-
with gr.Blocks(css='style.css') as demo:
|
272 |
gr.HTML(
|
273 |
"""
|
274 |
<div style="text-align: center; max-width: 1200px; margin: 20px auto;">
|
@@ -300,7 +261,7 @@ with gr.Blocks(css='style.css') as demo:
|
|
300 |
<img style="margin-top: 0em; margin-bottom: 0em" src="https://bit.ly/3gLdBN6" alt="Duplicate Space"></a>
|
301 |
</p>""")
|
302 |
|
303 |
-
with open('script.js', 'r') as f:
|
304 |
js_str = f.read()
|
305 |
|
306 |
demo.load(_js=js_str)
|
@@ -380,10 +341,10 @@ with gr.Blocks(css='style.css') as demo:
|
|
380 |
html_info = gr.HTML(elem_id=f'html_info', elem_classes="infotext")
|
381 |
|
382 |
inpaint_btn.click(
|
383 |
-
fn=
|
384 |
inputs=[
|
385 |
-
use_rasg,
|
386 |
model_picker,
|
|
|
387 |
use_painta,
|
388 |
prompt,
|
389 |
imageMask,
|
@@ -415,4 +376,4 @@ with gr.Blocks(css='style.css') as demo:
|
|
415 |
)
|
416 |
|
417 |
demo.queue(max_size=20)
|
418 |
-
demo.launch(share=True, allowed_paths=[TMP_DIR])
|
|
|
1 |
import os
|
2 |
+
import sys
|
3 |
+
from pathlib import Path
|
4 |
from collections import OrderedDict
|
5 |
|
6 |
import gradio as gr
|
7 |
import shutil
|
8 |
import uuid
|
9 |
import torch
|
|
|
|
|
10 |
from PIL import Image
|
11 |
|
12 |
+
demo_path = Path(__file__).resolve().parent
|
13 |
+
root_path = demo_path
|
14 |
+
sys.path.append(str(root_path))
|
15 |
+
from src import models
|
16 |
+
from src.methods import rasg, sd, sr
|
17 |
+
from src.utils import IImage, poisson_blend, image_from_url_text
|
18 |
|
19 |
|
20 |
+
TMP_DIR = root_path / 'gradio_tmp'
|
21 |
+
if TMP_DIR.exists():
|
22 |
+
shutil.rmtree(str(TMP_DIR))
|
23 |
+
TMP_DIR.mkdir(exist_ok=True, parents=True)
|
24 |
|
25 |
+
os.environ['GRADIO_TEMP_DIR'] = str(TMP_DIR)
|
26 |
|
27 |
on_huggingspace = os.environ.get("SPACE_AUTHOR_NAME") == "PAIR"
|
28 |
|
29 |
negative_prompt_str = "text, bad anatomy, bad proportions, blurry, cropped, deformed, disfigured, duplicate, error, extra limbs, gross proportions, jpeg artifacts, long neck, low quality, lowres, malformed, morbid, mutated, mutilated, out of frame, ugly, worst quality"
|
30 |
positive_prompt_str = "Full HD, 4K, high quality, high resolution"
|
31 |
|
32 |
+
examples_path = root_path / '__assets__/demo/examples'
|
33 |
example_inputs = [
|
34 |
+
[f'{examples_path}/images_1024/a40.jpg', f'{examples_path}/images_2048/a40.jpg', 'medieval castle'],
|
35 |
+
[f'{examples_path}/images_1024/a4.jpg', f'{examples_path}/images_2048/a4.jpg', 'parrot'],
|
36 |
+
[f'{examples_path}/images_1024/a65.jpg', f'{examples_path}/images_2048/a65.jpg', 'hoodie'],
|
37 |
+
[f'{examples_path}/images_1024/a54.jpg', f'{examples_path}/images_2048/a54.jpg', 'salad'],
|
38 |
+
[f'{examples_path}/images_1024/a51.jpg', f'{examples_path}/images_2048/a51.jpg', 'space helmet'],
|
39 |
+
[f'{examples_path}/images_1024/a46.jpg', f'{examples_path}/images_2048/a46.jpg', 'stack of books'],
|
40 |
+
[f'{examples_path}/images_1024/a19.jpg', f'{examples_path}/images_2048/a19.jpg', 'antique greek vase'],
|
41 |
+
[f'{examples_path}/images_1024/a2.jpg', f'{examples_path}/images_2048/a2.jpg', 'sunglasses'],
|
42 |
]
|
43 |
|
44 |
thumbnails = [
|
|
|
64 |
]
|
65 |
|
66 |
# Load models
|
67 |
+
models.pre_download_inpainting_models()
|
68 |
inpainting_models = OrderedDict([
|
69 |
+
("Dreamshaper Inpainting V8", 'ds8_inp'),
|
70 |
+
("Stable-Inpainting 2.0", 'sd2_inp'),
|
71 |
+
("Stable-Inpainting 1.5", 'sd15_inp')
|
72 |
])
|
73 |
sr_model = models.sd2_sr.load_model(device='cuda:1')
|
74 |
sam_predictor = models.sam.load_model(device='cuda:0')
|
75 |
|
76 |
+
inp_model_name = list(inpainting_models.keys())[0]
|
77 |
+
inp_model = models.load_inpainting_model(
|
78 |
+
inpainting_models[inp_model_name], device='cuda:0', cache=False)
|
79 |
+
|
80 |
+
|
81 |
+
def set_model_from_name(new_inp_model_name):
|
82 |
global inp_model
|
83 |
+
global inp_model_name
|
84 |
+
if new_inp_model_name != inp_model_name:
|
85 |
+
print (f"Activating Inpaintng Model: {new_inp_model_name}")
|
86 |
+
inp_model = models.load_inpainting_model(
|
87 |
+
inpainting_models[new_inp_model_name], device='cuda:0', cache=False)
|
88 |
+
inp_model_name = new_inp_model_name
|
89 |
|
90 |
|
91 |
def save_user_session(hr_image, hr_mask, lr_results, prompt, session_id=None):
|
92 |
if session_id == '':
|
93 |
session_id = str(uuid.uuid4())
|
94 |
|
95 |
+
session_dir = TMP_DIR / session_id
|
|
|
96 |
session_dir.mkdir(exist_ok=True, parents=True)
|
97 |
|
98 |
hr_image.save(session_dir / 'hr_image.png')
|
|
|
115 |
if session_id == '':
|
116 |
return None, None, [], ''
|
117 |
|
118 |
+
session_dir = TMP_DIR / session_id
|
|
|
119 |
lr_results_dir = session_dir / 'lr_results'
|
120 |
|
121 |
hr_image = Image.open(session_dir / 'hr_image.png')
|
|
|
132 |
return hr_image, hr_mask, gallery, prompt
|
133 |
|
134 |
|
135 |
+
def inpainting_run(model_name, use_rasg, use_painta, prompt, imageMask,
|
136 |
+
hr_image, seed, eta, negative_prompt, positive_prompt, ddim_steps,
|
137 |
+
guidance_scale=7.5, batch_size=1, session_id=''
|
|
|
|
|
138 |
):
|
139 |
torch.cuda.empty_cache()
|
140 |
+
set_model_from_name(model_name)
|
141 |
|
142 |
+
method = ['default']
|
|
|
|
|
|
|
|
|
|
|
|
|
143 |
if use_painta: method.append('painta')
|
144 |
+
if use_rasg: method.append('rasg')
|
145 |
method = '-'.join(method)
|
146 |
|
147 |
+
if use_rasg:
|
148 |
+
inpainting_f = rasg.run
|
149 |
+
else:
|
150 |
+
inpainting_f = sd.run
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
151 |
|
152 |
seed = int(seed)
|
153 |
batch_size = max(1, min(int(batch_size), 4))
|
|
|
164 |
for i in range(batch_size):
|
165 |
seed = seed + i * 1000
|
166 |
|
167 |
+
inpainted_image = inpainting_f(
|
168 |
ddim=inp_model,
|
169 |
method=method,
|
170 |
prompt=prompt,
|
|
|
195 |
|
196 |
def upscale_run(
|
197 |
ddim_steps, seed, use_sam_mask, session_id, img_index,
|
198 |
+
negative_prompt='', positive_prompt='high resolution professional photo'
|
|
|
199 |
):
|
200 |
hr_image, hr_mask, gallery, prompt = recover_user_session(session_id)
|
201 |
|
202 |
if len(gallery) == 0:
|
203 |
+
return Image.open(root_path / '__assets__/sr_info.png')
|
204 |
|
205 |
torch.cuda.empty_cache()
|
206 |
|
|
|
217 |
inpainted_image,
|
218 |
hr_image,
|
219 |
hr_mask,
|
220 |
+
prompt=f'{prompt}, {positive_prompt}',
|
221 |
noise_level=20,
|
222 |
blend_trick=True,
|
223 |
blend_output=True,
|
|
|
229 |
return output_image
|
230 |
|
231 |
|
232 |
+
with gr.Blocks(css=demo_path / 'style.css') as demo:
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
233 |
gr.HTML(
|
234 |
"""
|
235 |
<div style="text-align: center; max-width: 1200px; margin: 20px auto;">
|
|
|
261 |
<img style="margin-top: 0em; margin-bottom: 0em" src="https://bit.ly/3gLdBN6" alt="Duplicate Space"></a>
|
262 |
</p>""")
|
263 |
|
264 |
+
with open(demo_path / 'script.js', 'r') as f:
|
265 |
js_str = f.read()
|
266 |
|
267 |
demo.load(_js=js_str)
|
|
|
341 |
html_info = gr.HTML(elem_id=f'html_info', elem_classes="infotext")
|
342 |
|
343 |
inpaint_btn.click(
|
344 |
+
fn=inpainting_run,
|
345 |
inputs=[
|
|
|
346 |
model_picker,
|
347 |
+
use_rasg,
|
348 |
use_painta,
|
349 |
prompt,
|
350 |
imageMask,
|
|
|
376 |
)
|
377 |
|
378 |
demo.queue(max_size=20)
|
379 |
+
demo.launch(share=True, allowed_paths=[str(TMP_DIR)])
|
assets/.gitignore
DELETED
@@ -1 +0,0 @@
|
|
1 |
-
models/
|
|
|
|
config/ddpm/v1.yaml
ADDED
@@ -0,0 +1,14 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
linear_start: 0.00085
|
2 |
+
linear_end: 0.0120
|
3 |
+
num_timesteps_cond: 1
|
4 |
+
log_every_t: 200
|
5 |
+
timesteps: 1000
|
6 |
+
first_stage_key: "jpg"
|
7 |
+
cond_stage_key: "txt"
|
8 |
+
image_size: 64
|
9 |
+
channels: 4
|
10 |
+
cond_stage_trainable: false
|
11 |
+
conditioning_key: crossattn
|
12 |
+
monitor: val/loss_simple_ema
|
13 |
+
scale_factor: 0.18215
|
14 |
+
use_ema: False # we set this to false because this is an inference only config
|
config/ddpm/v2-upsample.yaml
ADDED
@@ -0,0 +1,24 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
parameterization: "v"
|
2 |
+
low_scale_key: "lr"
|
3 |
+
linear_start: 0.0001
|
4 |
+
linear_end: 0.02
|
5 |
+
num_timesteps_cond: 1
|
6 |
+
log_every_t: 200
|
7 |
+
timesteps: 1000
|
8 |
+
first_stage_key: "jpg"
|
9 |
+
cond_stage_key: "txt"
|
10 |
+
image_size: 128
|
11 |
+
channels: 4
|
12 |
+
cond_stage_trainable: false
|
13 |
+
conditioning_key: "hybrid-adm"
|
14 |
+
monitor: val/loss_simple_ema
|
15 |
+
scale_factor: 0.08333
|
16 |
+
use_ema: False
|
17 |
+
|
18 |
+
low_scale_config:
|
19 |
+
target: ldm.modules.diffusionmodules.upscaling.ImageConcatWithNoiseAugmentation
|
20 |
+
params:
|
21 |
+
noise_schedule_config: # image space
|
22 |
+
linear_start: 0.0001
|
23 |
+
linear_end: 0.02
|
24 |
+
max_noise_level: 350
|
config/encoders/clip.yaml
ADDED
@@ -0,0 +1 @@
|
|
|
|
|
1 |
+
__class__: smplfusion.models.encoders.clip_embedder.FrozenCLIPEmbedder
|
config/encoders/openclip.yaml
ADDED
@@ -0,0 +1,4 @@
|
|
|
|
|
|
|
|
|
|
|
1 |
+
__class__: smplfusion.models.encoders.open_clip_embedder.FrozenOpenCLIPEmbedder
|
2 |
+
__init__:
|
3 |
+
freeze: True
|
4 |
+
layer: "penultimate"
|
config/unet/inpainting/v1.yaml
ADDED
@@ -0,0 +1,15 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
__class__: smplfusion.models.unet.UNetModel
|
2 |
+
__init__:
|
3 |
+
image_size: 32 # unused
|
4 |
+
in_channels: 9 # 4 data + 4 downscaled image + 1 mask
|
5 |
+
out_channels: 4
|
6 |
+
model_channels: 320
|
7 |
+
attention_resolutions: [ 4, 2, 1 ]
|
8 |
+
num_res_blocks: 2
|
9 |
+
channel_mult: [ 1, 2, 4, 4 ]
|
10 |
+
num_heads: 8
|
11 |
+
use_spatial_transformer: True
|
12 |
+
transformer_depth: 1
|
13 |
+
context_dim: 768
|
14 |
+
use_checkpoint: False
|
15 |
+
legacy: False
|
config/unet/inpainting/v2.yaml
ADDED
@@ -0,0 +1,16 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
__class__: smplfusion.models.unet.UNetModel
|
2 |
+
__init__:
|
3 |
+
use_checkpoint: False
|
4 |
+
image_size: 32 # unused
|
5 |
+
in_channels: 9
|
6 |
+
out_channels: 4
|
7 |
+
model_channels: 320
|
8 |
+
attention_resolutions: [ 4, 2, 1 ]
|
9 |
+
num_res_blocks: 2
|
10 |
+
channel_mult: [ 1, 2, 4, 4 ]
|
11 |
+
num_head_channels: 64 # need to fix for flash-attn
|
12 |
+
use_spatial_transformer: True
|
13 |
+
use_linear_in_transformer: True
|
14 |
+
transformer_depth: 1
|
15 |
+
context_dim: 1024
|
16 |
+
legacy: False
|
config/unet/upsample/v2.yaml
ADDED
@@ -0,0 +1,19 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
__class__: smplfusion.models.unet.UNetModel
|
2 |
+
__init__:
|
3 |
+
use_checkpoint: False
|
4 |
+
num_classes: 1000 # timesteps for noise conditioning (here constant, just need one)
|
5 |
+
image_size: 128
|
6 |
+
in_channels: 7
|
7 |
+
out_channels: 4
|
8 |
+
model_channels: 256
|
9 |
+
attention_resolutions: [ 2,4,8]
|
10 |
+
num_res_blocks: 2
|
11 |
+
channel_mult: [ 1, 2, 2, 4]
|
12 |
+
disable_self_attentions: [True, True, True, False]
|
13 |
+
disable_middle_self_attn: False
|
14 |
+
num_heads: 8
|
15 |
+
use_spatial_transformer: True
|
16 |
+
transformer_depth: 1
|
17 |
+
context_dim: 1024
|
18 |
+
legacy: False
|
19 |
+
use_linear_in_transformer: True
|
config/vae-upsample.yaml
ADDED
@@ -0,0 +1,16 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
__class__: smplfusion.models.vae.AutoencoderKL
|
2 |
+
__init__:
|
3 |
+
embed_dim: 4
|
4 |
+
ddconfig:
|
5 |
+
double_z: True
|
6 |
+
z_channels: 4
|
7 |
+
resolution: 256
|
8 |
+
in_channels: 3
|
9 |
+
out_ch: 3
|
10 |
+
ch: 128
|
11 |
+
ch_mult: [ 1,2,4 ]
|
12 |
+
num_res_blocks: 2
|
13 |
+
attn_resolutions: [ ]
|
14 |
+
dropout: 0.0
|
15 |
+
lossconfig:
|
16 |
+
target: torch.nn.Identity
|
config/vae.yaml
ADDED
@@ -0,0 +1,17 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
__class__: smplfusion.models.vae.AutoencoderKL
|
2 |
+
__init__:
|
3 |
+
embed_dim: 4
|
4 |
+
monitor: val/rec_loss
|
5 |
+
ddconfig:
|
6 |
+
double_z: true
|
7 |
+
z_channels: 4
|
8 |
+
resolution: 256
|
9 |
+
in_channels: 3
|
10 |
+
out_ch: 3
|
11 |
+
ch: 128
|
12 |
+
ch_mult: [1,2,4,4]
|
13 |
+
num_res_blocks: 2
|
14 |
+
attn_resolutions: []
|
15 |
+
dropout: 0.0
|
16 |
+
lossconfig:
|
17 |
+
target: torch.nn.Identity
|
lib/models/__init__.py
DELETED
@@ -1 +0,0 @@
|
|
1 |
-
from . import sd2_inp, ds_inp, sd15_inp, sd2_sr, sam
|
|
|
|
lib/models/common.py
DELETED
@@ -1,49 +0,0 @@
|
|
1 |
-
import importlib
|
2 |
-
import requests
|
3 |
-
from pathlib import Path
|
4 |
-
from os.path import dirname
|
5 |
-
|
6 |
-
from omegaconf import OmegaConf
|
7 |
-
from tqdm import tqdm
|
8 |
-
|
9 |
-
|
10 |
-
PROJECT_DIR = dirname(dirname(dirname(__file__)))
|
11 |
-
CONFIG_FOLDER = f'{PROJECT_DIR}/assets/config'
|
12 |
-
MODEL_FOLDER = f'{PROJECT_DIR}/assets/models'
|
13 |
-
|
14 |
-
|
15 |
-
def download_file(url, save_path, chunk_size=1024):
|
16 |
-
try:
|
17 |
-
save_path = Path(save_path)
|
18 |
-
if save_path.exists():
|
19 |
-
print(f'{save_path.name} exists')
|
20 |
-
return
|
21 |
-
save_path.parent.mkdir(exist_ok=True, parents=True)
|
22 |
-
resp = requests.get(url, stream=True)
|
23 |
-
total = int(resp.headers.get('content-length', 0))
|
24 |
-
with open(save_path, 'wb') as file, tqdm(
|
25 |
-
desc=save_path.name,
|
26 |
-
total=total,
|
27 |
-
unit='iB',
|
28 |
-
unit_scale=True,
|
29 |
-
unit_divisor=1024,
|
30 |
-
) as bar:
|
31 |
-
for data in resp.iter_content(chunk_size=chunk_size):
|
32 |
-
size = file.write(data)
|
33 |
-
bar.update(size)
|
34 |
-
print(f'{save_path.name} download finished')
|
35 |
-
except Exception as e:
|
36 |
-
raise Exception(f"Download failed: {e}")
|
37 |
-
|
38 |
-
|
39 |
-
def get_obj_from_str(string):
|
40 |
-
module, cls = string.rsplit(".", 1)
|
41 |
-
try:
|
42 |
-
return getattr(importlib.import_module(module, package=None), cls)
|
43 |
-
except:
|
44 |
-
return getattr(importlib.import_module('lib.' + module, package=None), cls)
|
45 |
-
|
46 |
-
|
47 |
-
def load_obj(path):
|
48 |
-
objyaml = OmegaConf.load(path)
|
49 |
-
return get_obj_from_str(objyaml['__class__'])(**objyaml.get("__init__", {}))
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
lib/models/ds_inp.py
DELETED
@@ -1,51 +0,0 @@
|
|
1 |
-
import importlib
|
2 |
-
from omegaconf import OmegaConf
|
3 |
-
import torch
|
4 |
-
import safetensors
|
5 |
-
import safetensors.torch
|
6 |
-
|
7 |
-
from lib.smplfusion import DDIM, share, scheduler
|
8 |
-
from .common import *
|
9 |
-
|
10 |
-
|
11 |
-
MODEL_PATH = f'{MODEL_FOLDER}/dreamshaper/dreamshaper_8Inpainting.safetensors'
|
12 |
-
DOWNLOAD_URL = 'https://civitai.com/api/download/models/131004'
|
13 |
-
|
14 |
-
# pre-download
|
15 |
-
download_file(DOWNLOAD_URL, MODEL_PATH)
|
16 |
-
|
17 |
-
|
18 |
-
def load_model(dtype=torch.float16):
|
19 |
-
print ("Loading model: Dreamshaper Inpainting V8")
|
20 |
-
|
21 |
-
download_file(DOWNLOAD_URL, MODEL_PATH)
|
22 |
-
|
23 |
-
state_dict = safetensors.torch.load_file(MODEL_PATH)
|
24 |
-
|
25 |
-
config = OmegaConf.load(f'{CONFIG_FOLDER}/ddpm/v1.yaml')
|
26 |
-
unet = load_obj(f'{CONFIG_FOLDER}/unet/inpainting/v1.yaml').eval().cuda()
|
27 |
-
vae = load_obj(f'{CONFIG_FOLDER}/vae.yaml').eval().cuda()
|
28 |
-
encoder = load_obj(f'{CONFIG_FOLDER}/encoders/clip.yaml').eval().cuda()
|
29 |
-
|
30 |
-
extract = lambda state_dict, model: {x[len(model)+1:]:y for x,y in state_dict.items() if model in x}
|
31 |
-
unet_state = extract(state_dict, 'model.diffusion_model')
|
32 |
-
encoder_state = extract(state_dict, 'cond_stage_model')
|
33 |
-
vae_state = extract(state_dict, 'first_stage_model')
|
34 |
-
|
35 |
-
unet.load_state_dict(unet_state)
|
36 |
-
encoder.load_state_dict(encoder_state)
|
37 |
-
vae.load_state_dict(vae_state)
|
38 |
-
|
39 |
-
if dtype == torch.float16:
|
40 |
-
unet.convert_to_fp16()
|
41 |
-
vae.to(dtype)
|
42 |
-
encoder.to(dtype)
|
43 |
-
|
44 |
-
unet = unet.requires_grad_(False)
|
45 |
-
encoder = encoder.requires_grad_(False)
|
46 |
-
vae = vae.requires_grad_(False)
|
47 |
-
|
48 |
-
ddim = DDIM(config, vae, encoder, unet)
|
49 |
-
share.schedule = scheduler.linear(config.timesteps, config.linear_start, config.linear_end)
|
50 |
-
|
51 |
-
return ddim
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|