Update app.py
Browse files
app.py
CHANGED
@@ -191,26 +191,6 @@ title = """<h1 align="center">Mixture-of-Diffusers for SDXL Tiling Pipeline🤗<
|
|
191 |
</div>
|
192 |
"""
|
193 |
|
194 |
-
tips = """
|
195 |
-
### Method
|
196 |
-
The method proposed here strives to provide a better tool for image composition by using several diffusion processes in parallel, each configured with a specific prompt and settings, and focused on a particular region of the image. The mixture of diffusion processes is done in a way that harmonizes the generation process, preventing "seam" effects in the generated image.
|
197 |
-
Using several diffusion processes in parallel has also practical advantages when generating very large images, as the GPU memory requirements are similar to that of generating an image of the size of a single tile.
|
198 |
-
For practical demonstration purposes, this demo only covers image generation using 1x3 tiles. However, in the pipeline, you can freely increase the number of rows and columns as well as specify a row overlap.
|
199 |
-
|
200 |
-
### Tips
|
201 |
-
1. Describe the same environment for all image elements in your prompt. This helps to better harmonize the final image.
|
202 |
-
2. Keep the same stylization in both prompts.
|
203 |
-
3. Test different overlap sizes.
|
204 |
-
4. Test fews increments on seed.
|
205 |
-
5. This may take a while.
|
206 |
-
6. Enjoy!
|
207 |
-
"""
|
208 |
-
|
209 |
-
about = """
|
210 |
-
📧 **Contact**
|
211 |
-
<br>
|
212 |
-
If you have any questions or suggestions, feel free to send your question to <b>[email protected]</b>.
|
213 |
-
"""
|
214 |
|
215 |
with gr.Blocks(css=css) as app:
|
216 |
gr.Markdown(title)
|
|
|
191 |
</div>
|
192 |
"""
|
193 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
194 |
|
195 |
with gr.Blocks(css=css) as app:
|
196 |
gr.Markdown(title)
|