forked from s0md3v/sd-webui-roop
-
Notifications
You must be signed in to change notification settings - Fork 277
/
api_example.py
108 lines (96 loc) · 4.38 KB
/
api_example.py
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
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
import base64, io, requests, json
from PIL import Image, PngImagePlugin
from datetime import datetime, date
address = 'http://127.0.0.1:7860'
input_file = "extensions\sd-webui-reactor\example\IamSFW.jpg" # Input file path
time = datetime.now()
today = date.today()
current_date = today.strftime('%Y-%m-%d')
current_time = time.strftime('%H-%M-%S')
output = 'outputs/api/output_'+current_date+'_'+current_time # Output file path + name index
try:
im = Image.open(input_file)
except Exception as e:
print(e)
finally:
print(im)
img_bytes = io.BytesIO()
im.save(img_bytes, format='PNG')
img_base64 = base64.b64encode(img_bytes.getvalue()).decode('utf-8')
# ReActor arguments:
args=[
img_base64, #0
True, #1 Enable ReActor
'0', #2 Comma separated face number(s) from swap-source image
'0', #3 Comma separated face number(s) for target image (result)
'C:\stable-diffusion-webui\models\insightface\inswapper_128.onnx', #4 model path
'CodeFormer', #4 Restore Face: None; CodeFormer; GFPGAN
1, #5 Restore visibility value
True, #7 Restore face -> Upscale
'4x_NMKD-Superscale-SP_178000_G', #8 Upscaler (type 'None' if doesn't need), see full list here: http://127.0.0.1:7860/sdapi/v1/script-info -> reactor -> sec.8
1.5, #9 Upscaler scale value
1, #10 Upscaler visibility (if scale = 1)
False, #11 Swap in source image
True, #12 Swap in generated image
1, #13 Console Log Level (0 - min, 1 - med or 2 - max)
0, #14 Gender Detection (Source) (0 - No, 1 - Female Only, 2 - Male Only)
0, #15 Gender Detection (Target) (0 - No, 1 - Female Only, 2 - Male Only)
False, #16 Save the original image(s) made before swapping
0.8, #17 CodeFormer Weight (0 = maximum effect, 1 = minimum effect), 0.5 - by default
False, #18 Source Image Hash Check, True - by default
False, #19 Target Image Hash Check, False - by default
"CUDA", #20 CPU or CUDA (if you have it), CPU - by default
True, #21 Face Mask Correction
1, #22 Select Source, 0 - Image, 1 - Face Model, 2 - Source Folder
"elena.safetensors", #23 Filename of the face model (from "models/reactor/faces"), e.g. elena.safetensors, don't forger to set #22 to 1
"C:\PATH_TO_FACES_IMAGES", #24 The path to the folder containing source faces images, don't forger to set #22 to 2
None, #25 skip it for API
True, #26 Randomly select an image from the path
True, #27 Force Upscale even if no face found
0.6, #28 Face Detection Threshold
2, #29 Maximum number of faces to detect (0 is unlimited)
]
# The args for ReActor can be found by
# requests.get(url=f'{address}/sdapi/v1/script-info')
prompt = "(8k, best quality, masterpiece, highly detailed:1.1),realistic photo of fantastic happy woman,hairstyle of blonde and red short bob hair,modern clothing,cinematic lightning,film grain,dynamic pose,bokeh,dof"
neg = "ng_deepnegative_v1_75t,(badhandv4:1.2),(worst quality:2),(low quality:2),(normal quality:2),lowres,(bad anatomy),(bad hands),((monochrome)),((grayscale)),(verybadimagenegative_v1.3:0.8),negative_hand-neg,badhandv4,nude,naked,(strabismus),cross-eye,heterochromia,((blurred))"
payload = {
"prompt": prompt,
"negative_prompt": neg,
"seed": -1,
"sampler_name": "DPM++ 2M Karras",
"steps": 15,
"cfg_scale": 7,
"width": 512,
"height": 768,
"restore_faces": False,
"alwayson_scripts": {"reactor":{"args":args}}
}
try:
print('Working... Please wait...')
result = requests.post(url=f'{address}/sdapi/v1/txt2img', json=payload)
except Exception as e:
print(e)
finally:
print('Done! Saving file...')
if result is not None:
r = result.json()
n = 0
for i in r['images']:
image = Image.open(io.BytesIO(base64.b64decode(i.split(",",1)[0])))
png_payload = {
"image": "data:image/png;base64," + i
}
response2 = requests.post(url=f'{address}/sdapi/v1/png-info', json=png_payload)
pnginfo = PngImagePlugin.PngInfo()
pnginfo.add_text("parameters", response2.json().get("info"))
output_file = output+'_'+str(n)+'_.png'
try:
image.save(output_file, pnginfo=pnginfo)
except Exception as e:
print(e)
finally:
print(f'{output_file} is saved\nAll is done!')
n += 1
else:
print('Something went wrong...')