Spaces:
Paused
Paused
ResearcherXman
commited on
Commit
•
3c49f9a
1
Parent(s):
4912220
update
Browse files- app.py +195 -4
- requirements.txt +14 -0
app.py
CHANGED
@@ -1,7 +1,198 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
import gradio as gr
|
2 |
|
3 |
-
|
4 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
5 |
|
6 |
-
|
7 |
-
|
|
|
|
1 |
+
import os
|
2 |
+
import cv2
|
3 |
+
import math
|
4 |
+
import random
|
5 |
+
import numpy as np
|
6 |
+
from PIL import Image
|
7 |
+
|
8 |
+
from diffusers.utils import load_image
|
9 |
+
|
10 |
import gradio as gr
|
11 |
|
12 |
+
# global variable
|
13 |
+
MAX_SEED = np.iinfo(np.int32).max
|
14 |
+
|
15 |
+
def randomize_seed_fn(seed: int, randomize_seed: bool) -> int:
|
16 |
+
if randomize_seed:
|
17 |
+
seed = random.randint(0, MAX_SEED)
|
18 |
+
return seed
|
19 |
+
|
20 |
+
def swap_to_gallery(images):
|
21 |
+
return gr.update(value=images, visible=True), gr.update(visible=True), gr.update(visible=False)
|
22 |
+
|
23 |
+
def upload_example_to_gallery(images, prompt, style, negative_prompt):
|
24 |
+
return gr.update(value=images, visible=True), gr.update(visible=True), gr.update(visible=False)
|
25 |
+
|
26 |
+
def remove_back_to_files():
|
27 |
+
return gr.update(visible=False), gr.update(visible=False), gr.update(visible=True)
|
28 |
+
|
29 |
+
def remove_tips():
|
30 |
+
return gr.update(visible=False)
|
31 |
+
|
32 |
+
def generate_image(face_image, pose_image, prompt, negative_prompt, num_steps, identitynet_strength_ratio, adapter_strength_ratio, num_outputs, guidance_scale, seed, progress=gr.Progress(track_tqdm=True)):
|
33 |
+
|
34 |
+
if face_image is None:
|
35 |
+
raise gr.Error(f"Cannot find any input face image! Please upload the face image")
|
36 |
+
|
37 |
+
face_image = load_image(face_image[0])
|
38 |
+
|
39 |
+
return [face_image], gr.update(visible=True)
|
40 |
+
|
41 |
+
### Description
|
42 |
+
title = r"""
|
43 |
+
<h1 align="center">InstantID: Zero-shot Identity-Preserving Generation in Seconds</h1>
|
44 |
+
"""
|
45 |
+
|
46 |
+
description = r"""
|
47 |
+
<b>Official 🤗 Gradio demo</b> for <a href='https://github.com/InstantID/InstantID' target='_blank'><b>InstantID: Zero-shot Identity-Preserving Generation in Seconds</b></a>.<br>
|
48 |
+
|
49 |
+
How to use:<br>
|
50 |
+
1. Upload a person image or cropped face image. For multiple person images, we will only detect the biggest face. Make sure face is in good condition and not significantly blocked or blurred.
|
51 |
+
2. (Optionally) upload another person image as reference pose. If not uploaded, we will use the first person image to extract landmarks.
|
52 |
+
3. Enter a text prompt as normal text-to-image model.
|
53 |
+
4. Click the <b>Submit</b> button to start customizing.
|
54 |
+
5. Share your customizd photo with your friends, enjoy😊!
|
55 |
+
"""
|
56 |
+
|
57 |
+
article = r"""
|
58 |
+
---
|
59 |
+
📝 **Citation**
|
60 |
+
<br>
|
61 |
+
If our work is helpful for your research or applications, please cite us via:
|
62 |
+
```bibtex
|
63 |
+
@misc{wang2024instantid,
|
64 |
+
title={InstantID: Zero-shot Identity-Preserving Generation in Seconds},
|
65 |
+
author={Qixun Wang and Xu Bai and Haofan Wang and Zekui Qin and Anthony Chen},
|
66 |
+
year={2024},
|
67 |
+
eprint={2401.07519},
|
68 |
+
archivePrefix={arXiv},
|
69 |
+
primaryClass={cs.CV}
|
70 |
+
}
|
71 |
+
```
|
72 |
+
|
73 |
+
📧 **Contact**
|
74 |
+
<br>
|
75 |
+
If you have any questions, please feel free to open an issue or directly reach us out at <b>[email protected]</b>.
|
76 |
+
"""
|
77 |
+
|
78 |
+
tips = r"""
|
79 |
+
### Usage tips of InstantID
|
80 |
+
1. If you're not satisfied with the similarity, scroll down to "Advanced Options" and increase the weight of "IdentityNet Strength" and "Adapter Strength".
|
81 |
+
2. If you feel that the saturation is too high, first decrease the Adapter strength. If it is still too high, then decrease the IdentityNet strength.
|
82 |
+
3. If you find that text control is not as expected, decrease Adapter strength.
|
83 |
+
"""
|
84 |
+
|
85 |
+
css = '''
|
86 |
+
.gradio-container {width: 85% !important}
|
87 |
+
'''
|
88 |
+
with gr.Blocks(css=css) as demo:
|
89 |
+
|
90 |
+
# description
|
91 |
+
gr.Markdown(title)
|
92 |
+
gr.Markdown(description)
|
93 |
+
|
94 |
+
with gr.Row():
|
95 |
+
with gr.Column():
|
96 |
+
|
97 |
+
# upload face image
|
98 |
+
face_files = gr.Files(
|
99 |
+
label="Upload a photo of your face",
|
100 |
+
file_types=["image"]
|
101 |
+
)
|
102 |
+
uploaded_faces = gr.Gallery(label="Your images", visible=False, columns=1, rows=1, height=512)
|
103 |
+
with gr.Column(visible=False) as clear_button_face:
|
104 |
+
remove_and_reupload_faces = gr.ClearButton(value="Remove and upload new ones", components=face_files, size="sm")
|
105 |
+
|
106 |
+
# optional: upload a reference pose image
|
107 |
+
pose_files = gr.Files(
|
108 |
+
label="Upload a reference pose image (optional)",
|
109 |
+
file_types=["image"]
|
110 |
+
)
|
111 |
+
uploaded_poses = gr.Gallery(label="Your images", visible=False, columns=1, rows=1, height=512)
|
112 |
+
with gr.Column(visible=False) as clear_button_pose:
|
113 |
+
remove_and_reupload_poses = gr.ClearButton(value="Remove and upload new ones", components=pose_files, size="sm")
|
114 |
+
|
115 |
+
# prompt
|
116 |
+
prompt = gr.Textbox(label="Prompt",
|
117 |
+
info="Give simple prompt is enough to achieve good face fedility",
|
118 |
+
placeholder="A photo of a man/woman")
|
119 |
+
submit = gr.Button("Submit")
|
120 |
+
|
121 |
+
with gr.Accordion(open=False, label="Advanced Options"):
|
122 |
+
negative_prompt = gr.Textbox(
|
123 |
+
label="Negative Prompt",
|
124 |
+
placeholder="low quality",
|
125 |
+
value="nsfw, lowres, bad anatomy, bad hands, text, error, missing fingers, extra digit, fewer digits, cropped, worst quality, low quality, normal quality, jpeg artifacts, signature, watermark, username, blurry",
|
126 |
+
)
|
127 |
+
num_steps = gr.Slider(
|
128 |
+
label="Number of sample steps",
|
129 |
+
minimum=20,
|
130 |
+
maximum=100,
|
131 |
+
step=1,
|
132 |
+
value=30,
|
133 |
+
)
|
134 |
+
identitynet_strength_ratio = gr.Slider(
|
135 |
+
label="IdentityNet strength",
|
136 |
+
minimum=0,
|
137 |
+
maximum=1.5,
|
138 |
+
step=0.05,
|
139 |
+
value=0.65,
|
140 |
+
)
|
141 |
+
adapter_strength_ratio = gr.Slider(
|
142 |
+
label="Image adapter strength",
|
143 |
+
minimum=0,
|
144 |
+
maximum=1,
|
145 |
+
step=0.05,
|
146 |
+
value=0.30,
|
147 |
+
)
|
148 |
+
num_outputs = gr.Slider(
|
149 |
+
label="Number of output images",
|
150 |
+
minimum=1,
|
151 |
+
maximum=4,
|
152 |
+
step=1,
|
153 |
+
value=2,
|
154 |
+
)
|
155 |
+
guidance_scale = gr.Slider(
|
156 |
+
label="Guidance scale",
|
157 |
+
minimum=0.1,
|
158 |
+
maximum=10.0,
|
159 |
+
step=0.1,
|
160 |
+
value=5,
|
161 |
+
)
|
162 |
+
seed = gr.Slider(
|
163 |
+
label="Seed",
|
164 |
+
minimum=0,
|
165 |
+
maximum=MAX_SEED,
|
166 |
+
step=1,
|
167 |
+
value=42,
|
168 |
+
)
|
169 |
+
randomize_seed = gr.Checkbox(label="Randomize seed", value=True)
|
170 |
+
|
171 |
+
with gr.Column():
|
172 |
+
gallery = gr.Gallery(label="Generated Images")
|
173 |
+
usage_tips = gr.Markdown(label="Usage tips of InstantID", value=tips ,visible=False)
|
174 |
+
|
175 |
+
face_files.upload(fn=swap_to_gallery, inputs=face_files, outputs=[uploaded_faces, clear_button_face, face_files])
|
176 |
+
pose_files.upload(fn=swap_to_gallery, inputs=pose_files, outputs=[uploaded_poses, clear_button_pose, pose_files])
|
177 |
+
|
178 |
+
remove_and_reupload_faces.click(fn=remove_back_to_files, outputs=[uploaded_faces, clear_button_face, face_files])
|
179 |
+
remove_and_reupload_poses.click(fn=remove_back_to_files, outputs=[uploaded_poses, clear_button_pose, pose_files])
|
180 |
+
|
181 |
+
submit.click(
|
182 |
+
fn=remove_tips,
|
183 |
+
outputs=usage_tips,
|
184 |
+
).then(
|
185 |
+
fn=randomize_seed_fn,
|
186 |
+
inputs=[seed, randomize_seed],
|
187 |
+
outputs=seed,
|
188 |
+
queue=False,
|
189 |
+
api_name=False,
|
190 |
+
).then(
|
191 |
+
fn=generate_image,
|
192 |
+
inputs=[face_files, pose_files, prompt, negative_prompt, num_steps, identitynet_strength_ratio, adapter_strength_ratio, num_outputs, guidance_scale, seed],
|
193 |
+
outputs=[gallery, usage_tips]
|
194 |
+
)
|
195 |
|
196 |
+
gr.Markdown(article)
|
197 |
+
|
198 |
+
demo.launch()
|
requirements.txt
ADDED
@@ -0,0 +1,14 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
diffusers==0.25.0
|
2 |
+
torch==2.0.0
|
3 |
+
torchvision==0.15.1
|
4 |
+
transformers==4.36.2
|
5 |
+
accelerate
|
6 |
+
safetensors
|
7 |
+
einops
|
8 |
+
onnxruntime-gpu
|
9 |
+
spaces==0.19.4
|
10 |
+
omegaconf
|
11 |
+
peft
|
12 |
+
huggingface-hub==0.20.2
|
13 |
+
opencv-python
|
14 |
+
insightface
|