Spaces:
Paused
Paused
Update app.py
Browse files
app.py
CHANGED
@@ -128,16 +128,15 @@ def get_example():
|
|
128 |
|
129 |
### Description and style
|
130 |
logo = r"""
|
131 |
-
<center><img src='https://
|
132 |
"""
|
133 |
title = r"""
|
134 |
-
<h1 align="center">
|
135 |
"""
|
136 |
|
137 |
description = r"""
|
138 |
-
<b>Official
|
139 |
<br>
|
140 |
-
For stylization, you could use our other gradio demo [PhotoMaker-Style](https://huggingface.co/spaces/TencentARC/PhotoMaker-Style).
|
141 |
<br>
|
142 |
❗️❗️❗️[<b>Important</b>] Personalization steps:<br>
|
143 |
1️⃣ Upload images of someone you want to customize. One image is ok, but more is better. Although we do not perform face detection, the face in the uploaded image should <b>occupy the majority of the image</b>.<br>
|
@@ -148,20 +147,10 @@ For stylization, you could use our other gradio demo [PhotoMaker-Style](https://
|
|
148 |
|
149 |
article = r"""
|
150 |
|
151 |
-
If
|
152 |
-
|
153 |
---
|
154 |
-
|
155 |
-
<br>
|
156 |
-
If our work is useful for your research, please consider citing:
|
157 |
-
|
158 |
-
```bibtex
|
159 |
-
@article{li2023photomaker,
|
160 |
-
title={PhotoMaker: Customizing Realistic Human Photos via Stacked ID Embedding},
|
161 |
-
author={Li, Zhen and Cao, Mingdeng and Wang, Xintao and Qi, Zhongang and Cheng, Ming-Ming and Shan, Ying},
|
162 |
-
booktitle={arXiv preprint arxiv:2312.04461},
|
163 |
-
year={2023}
|
164 |
-
}
|
165 |
```
|
166 |
📋 **License**
|
167 |
<br>
|
@@ -169,14 +158,13 @@ Apache-2.0 LICENSE. Please refer to the [LICENSE file](https://huggingface.co/Te
|
|
169 |
|
170 |
📧 **Contact**
|
171 |
<br>
|
172 |
-
If you have any questions, please feel free to reach me out at <b>
|
173 |
"""
|
174 |
|
175 |
tips = r"""
|
176 |
-
### Usage tips of
|
177 |
1. Upload more photos of the person to be customized to **improve ID fidelty**. If the input is Asian face(s), maybe consider adding 'asian' before the class word, e.g., `asian woman img`
|
178 |
-
2.
|
179 |
-
3. For **faster** speed, reduce the number of generated images and sampling steps. However, please note that reducing the sampling steps may compromise the ID fidelity.
|
180 |
"""
|
181 |
# We have provided some generate examples and comparisons at: [this website]().
|
182 |
# 3. Don't make the prompt too long, as we will trim it if it exceeds 77 tokens.
|
|
|
128 |
|
129 |
### Description and style
|
130 |
logo = r"""
|
131 |
+
<center><img src='https://img.mytsi.org/i/1eyo451.png' alt='PhotoMaker logo' style="width:80px; margin-bottom:10px"></center>
|
132 |
"""
|
133 |
title = r"""
|
134 |
+
<h1 align="center">Pixio Photo: Customizing Realistic Human Photos via Stacked ID Embedding</h1>
|
135 |
"""
|
136 |
|
137 |
description = r"""
|
138 |
+
<b>Official Demo</b> DM feedback to <a href='https://twitter.com/tsi_org' target='_blank'><b>Pixio Photo: Customizing Realistic Human Photos via Stacked ID Embedding</b></a>.<br>
|
139 |
<br>
|
|
|
140 |
<br>
|
141 |
❗️❗️❗️[<b>Important</b>] Personalization steps:<br>
|
142 |
1️⃣ Upload images of someone you want to customize. One image is ok, but more is better. Although we do not perform face detection, the face in the uploaded image should <b>occupy the majority of the image</b>.<br>
|
|
|
147 |
|
148 |
article = r"""
|
149 |
|
150 |
+
If Pixio Photo is helpful, please help to DM the <a href='https://twitter.com/tsi_org' target='_blank'>Github Repo</a>. Thanks!
|
151 |
+
|
152 |
---
|
153 |
+
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
154 |
```
|
155 |
📋 **License**
|
156 |
<br>
|
|
|
158 |
|
159 |
📧 **Contact**
|
160 |
<br>
|
161 |
+
If you have any questions, please feel free to reach me out at <b>rcohen@mytsi.org</b>.
|
162 |
"""
|
163 |
|
164 |
tips = r"""
|
165 |
+
### Usage tips of Pixio Photo
|
166 |
1. Upload more photos of the person to be customized to **improve ID fidelty**. If the input is Asian face(s), maybe consider adding 'asian' before the class word, e.g., `asian woman img`
|
167 |
+
2. For **faster** speed, reduce the number of generated images and sampling steps. However, please note that reducing the sampling steps may compromise the ID fidelity.
|
|
|
168 |
"""
|
169 |
# We have provided some generate examples and comparisons at: [this website]().
|
170 |
# 3. Don't make the prompt too long, as we will trim it if it exceeds 77 tokens.
|