Spaces:
Running
on
Zero
Running
on
Zero
zhiweili
commited on
Commit
•
991068d
1
Parent(s):
991954d
change to img2img
Browse files- app.py +1 -1
- app_haircolor_img2img.py +40 -37
app.py
CHANGED
@@ -1,6 +1,6 @@
|
|
1 |
import gradio as gr
|
2 |
|
3 |
-
from
|
4 |
|
5 |
with gr.Blocks(css="style.css") as demo:
|
6 |
with gr.Tabs():
|
|
|
1 |
import gradio as gr
|
2 |
|
3 |
+
from app_haircolor_img2img import create_demo as create_demo_haircolor
|
4 |
|
5 |
with gr.Blocks(css="style.css") as demo:
|
6 |
with gr.Tabs():
|
app_haircolor_img2img.py
CHANGED
@@ -10,22 +10,19 @@ from segment_utils import(
|
|
10 |
restore_result,
|
11 |
)
|
12 |
from diffusers import (
|
13 |
-
|
14 |
-
|
15 |
-
|
16 |
-
|
17 |
EulerAncestralDiscreteScheduler,
|
18 |
-
UniPCMultistepScheduler,
|
19 |
)
|
20 |
|
21 |
from controlnet_aux import (
|
22 |
CannyDetector,
|
23 |
LineartDetector,
|
24 |
-
PidiNetDetector,
|
25 |
-
HEDdetector,
|
26 |
)
|
27 |
|
28 |
-
BASE_MODEL = "stable-diffusion-
|
29 |
|
30 |
DEVICE = "cuda" if torch.cuda.is_available() else "cpu"
|
31 |
|
@@ -37,32 +34,34 @@ canny_detector = CannyDetector()
|
|
37 |
lineart_detector = LineartDetector.from_pretrained("lllyasviel/Annotators")
|
38 |
lineart_detector = lineart_detector.to(DEVICE)
|
39 |
|
40 |
-
pidiNet_detector = PidiNetDetector.from_pretrained('lllyasviel/Annotators')
|
41 |
-
pidiNet_detector = pidiNet_detector.to(DEVICE)
|
42 |
|
43 |
-
|
44 |
-
|
45 |
-
|
46 |
-
|
47 |
-
|
48 |
-
|
49 |
-
|
50 |
-
|
51 |
-
|
52 |
-
|
53 |
-
|
54 |
-
|
55 |
-
]
|
|
|
|
|
56 |
|
57 |
-
basepipeline =
|
58 |
BASE_MODEL,
|
59 |
-
torch_dtype=torch.float16,
|
|
|
60 |
use_safetensors=True,
|
61 |
-
|
|
|
|
|
|
|
62 |
)
|
63 |
|
64 |
-
basepipeline.scheduler = UniPCMultistepScheduler.from_config(basepipeline.scheduler.config)
|
65 |
-
|
66 |
basepipeline = basepipeline.to(DEVICE)
|
67 |
|
68 |
basepipeline.enable_model_cpu_offload()
|
@@ -78,15 +77,17 @@ def image_to_image(
|
|
78 |
generate_size: int,
|
79 |
cond_scale1: float = 1.2,
|
80 |
cond_scale2: float = 1.2,
|
|
|
|
|
81 |
):
|
82 |
run_task_time = 0
|
83 |
time_cost_str = ''
|
84 |
run_task_time, time_cost_str = get_time_cost(run_task_time, time_cost_str)
|
85 |
-
lineart_image = lineart_detector(input_image,
|
|
|
86 |
|
87 |
-
|
88 |
-
|
89 |
-
cond_image = [lineart_image, pidinet_image]
|
90 |
|
91 |
generator = torch.Generator(device=DEVICE).manual_seed(seed)
|
92 |
generated_image = basepipeline(
|
@@ -99,7 +100,7 @@ def image_to_image(
|
|
99 |
guidance_scale=guidance_scale,
|
100 |
strength=strength,
|
101 |
num_inference_steps=num_steps,
|
102 |
-
controlnet_conditioning_scale=
|
103 |
).images[0]
|
104 |
|
105 |
run_task_time, time_cost_str = get_time_cost(run_task_time, time_cost_str)
|
@@ -137,15 +138,17 @@ def create_demo() -> gr.Blocks:
|
|
137 |
with gr.Column():
|
138 |
num_steps = gr.Slider(minimum=1, maximum=100, value=20, step=1, label="Num Steps")
|
139 |
guidance_scale = gr.Slider(minimum=0, maximum=30, value=5, step=0.5, label="Guidance Scale")
|
140 |
-
strength = gr.Slider(minimum=0, maximum=3, value=0.2, step=0.1, label="Strength")
|
141 |
with gr.Column():
|
|
|
142 |
with gr.Accordion("Advanced Options", open=False):
|
143 |
mask_expansion = gr.Number(label="Mask Expansion", value=50, visible=True)
|
144 |
mask_dilation = gr.Slider(minimum=0, maximum=10, value=2, step=1, label="Mask Dilation")
|
145 |
seed = gr.Number(label="Seed", value=8)
|
146 |
category = gr.Textbox(label="Category", value=DEFAULT_CATEGORY, visible=False)
|
147 |
-
cond_scale1 = gr.Slider(minimum=0, maximum=3, value=
|
148 |
-
cond_scale2 = gr.Slider(minimum=0, maximum=3, value=
|
|
|
|
|
149 |
g_btn = gr.Button("Edit Image")
|
150 |
|
151 |
with gr.Row():
|
@@ -164,7 +167,7 @@ def create_demo() -> gr.Blocks:
|
|
164 |
outputs=[origin_area_image, croper],
|
165 |
).success(
|
166 |
fn=image_to_image,
|
167 |
-
inputs=[origin_area_image, edit_prompt,seed, num_steps, guidance_scale, strength, generate_size, cond_scale1, cond_scale2],
|
168 |
outputs=[generated_image, generated_cost],
|
169 |
).success(
|
170 |
fn=restore_result,
|
|
|
10 |
restore_result,
|
11 |
)
|
12 |
from diffusers import (
|
13 |
+
DiffusionPipeline,
|
14 |
+
T2IAdapter,
|
15 |
+
MultiAdapter,
|
16 |
+
AutoencoderKL,
|
17 |
EulerAncestralDiscreteScheduler,
|
|
|
18 |
)
|
19 |
|
20 |
from controlnet_aux import (
|
21 |
CannyDetector,
|
22 |
LineartDetector,
|
|
|
|
|
23 |
)
|
24 |
|
25 |
+
BASE_MODEL = "stabilityai/stable-diffusion-xl-base-1.0"
|
26 |
|
27 |
DEVICE = "cuda" if torch.cuda.is_available() else "cpu"
|
28 |
|
|
|
34 |
lineart_detector = LineartDetector.from_pretrained("lllyasviel/Annotators")
|
35 |
lineart_detector = lineart_detector.to(DEVICE)
|
36 |
|
|
|
|
|
37 |
|
38 |
+
adapters = MultiAdapter(
|
39 |
+
[
|
40 |
+
T2IAdapter.from_pretrained(
|
41 |
+
"TencentARC/t2i-adapter-lineart-sdxl-1.0",
|
42 |
+
torch_dtype=torch.float16,
|
43 |
+
varient="fp16",
|
44 |
+
),
|
45 |
+
T2IAdapter.from_pretrained(
|
46 |
+
"TencentARC/t2i-adapter-canny-sdxl-1.0",
|
47 |
+
torch_dtype=torch.float16,
|
48 |
+
varient="fp16",
|
49 |
+
),
|
50 |
+
]
|
51 |
+
)
|
52 |
+
adapters = adapters.to(torch.float16)
|
53 |
|
54 |
+
basepipeline = DiffusionPipeline.from_pretrained(
|
55 |
BASE_MODEL,
|
56 |
+
torch_dtype=torch.float16,
|
57 |
+
variant="fp16",
|
58 |
use_safetensors=True,
|
59 |
+
vae=AutoencoderKL.from_pretrained("madebyollin/sdxl-vae-fp16-fix", torch_dtype=torch.float16),
|
60 |
+
scheduler=EulerAncestralDiscreteScheduler.from_pretrained(BASE_MODEL, subfolder="scheduler"),
|
61 |
+
adapter=adapters,
|
62 |
+
custom_pipeline="./pipelines/pipeline_sdxl_adapter_img2img.py",
|
63 |
)
|
64 |
|
|
|
|
|
65 |
basepipeline = basepipeline.to(DEVICE)
|
66 |
|
67 |
basepipeline.enable_model_cpu_offload()
|
|
|
77 |
generate_size: int,
|
78 |
cond_scale1: float = 1.2,
|
79 |
cond_scale2: float = 1.2,
|
80 |
+
lineart_detect:float = 0.375,
|
81 |
+
canny_detect:float = 0.375,
|
82 |
):
|
83 |
run_task_time = 0
|
84 |
time_cost_str = ''
|
85 |
run_task_time, time_cost_str = get_time_cost(run_task_time, time_cost_str)
|
86 |
+
lineart_image = lineart_detector(input_image, int(generate_size * lineart_detect), generate_size)
|
87 |
+
canny_image = canny_detector(input_image, int(generate_size * canny_detect), generate_size)
|
88 |
|
89 |
+
cond_image = [lineart_image, canny_image]
|
90 |
+
cond_scale = [cond_scale1, cond_scale2]
|
|
|
91 |
|
92 |
generator = torch.Generator(device=DEVICE).manual_seed(seed)
|
93 |
generated_image = basepipeline(
|
|
|
100 |
guidance_scale=guidance_scale,
|
101 |
strength=strength,
|
102 |
num_inference_steps=num_steps,
|
103 |
+
controlnet_conditioning_scale=cond_scale,
|
104 |
).images[0]
|
105 |
|
106 |
run_task_time, time_cost_str = get_time_cost(run_task_time, time_cost_str)
|
|
|
138 |
with gr.Column():
|
139 |
num_steps = gr.Slider(minimum=1, maximum=100, value=20, step=1, label="Num Steps")
|
140 |
guidance_scale = gr.Slider(minimum=0, maximum=30, value=5, step=0.5, label="Guidance Scale")
|
|
|
141 |
with gr.Column():
|
142 |
+
strength = gr.Slider(minimum=0, maximum=3, value=0.2, step=0.1, label="Strength")
|
143 |
with gr.Accordion("Advanced Options", open=False):
|
144 |
mask_expansion = gr.Number(label="Mask Expansion", value=50, visible=True)
|
145 |
mask_dilation = gr.Slider(minimum=0, maximum=10, value=2, step=1, label="Mask Dilation")
|
146 |
seed = gr.Number(label="Seed", value=8)
|
147 |
category = gr.Textbox(label="Category", value=DEFAULT_CATEGORY, visible=False)
|
148 |
+
cond_scale1 = gr.Slider(minimum=0, maximum=3, value=0.8, step=0.1, label="Cond_scale1")
|
149 |
+
cond_scale2 = gr.Slider(minimum=0, maximum=3, value=0.3, step=0.1, label="Cond_scale2")
|
150 |
+
lineart_detect = gr.Slider(minimum=0, maximum=1, value=0.375, step=0.01, label="Lineart Detect")
|
151 |
+
canny_detect = gr.Slider(minimum=0, maximum=1, value=0.75, step=0.01, label="Canny Detect")
|
152 |
g_btn = gr.Button("Edit Image")
|
153 |
|
154 |
with gr.Row():
|
|
|
167 |
outputs=[origin_area_image, croper],
|
168 |
).success(
|
169 |
fn=image_to_image,
|
170 |
+
inputs=[origin_area_image, edit_prompt,seed, num_steps, guidance_scale, strength, generate_size, cond_scale1, cond_scale2, lineart_detect, canny_detect],
|
171 |
outputs=[generated_image, generated_cost],
|
172 |
).success(
|
173 |
fn=restore_result,
|