I was going to write it in my free time before yesterday&#39;s dinner, but I couldn&#39;t, so I carried it over.
 - <a href="/en/Seed%20and%20prompt%20relationship%20for%20Stable%20Diffusion">Seed and prompt relationship for Stable Diffusion</a>
<ul>
<li>About inpaint
<ul>
<li><blockquote>
It is easier to programmatically create the mask image itself by looking at the differences.
</blockquote>
<ul>
<li><blockquote>
Prepare the original image and the image painted with a semi-transparent marker, and mask the areas where there are differences.
</blockquote>
</li>
</ul>
</li>
<li>This didn&#39;t work.<ul>
<li>I sent the PNG to my iPad, added it back, and it became a JPEG😅.</li>
<li>Should I open and edit with a decent drawing tool instead of the standard addendum tool?</li>
<li>But if that&#39;s the case, why not just use layers in the first place?</li>
</ul>
</li>
</ul>
</li>
<li><a href="/en/Diary%202022-09-26">Diary 2022-09-26</a> ←Diary 2022-09-27 → <a href="/en/Diary%202022-09-28">Diary 2022-09-28</a>
100 days ago <a href="/en/Diary%202022-06-19">Diary 2022-06-19</a>.
1 year ago <a href="/en/Diary%202021-09-27">Diary 2021-09-27</a>.
</li>
</ul>
<hr>
This page is auto-translated from <a href="https://scrapbox.io/nishio/%E6%97%A5%E8%A8%982022-09-27">/nishio/日記2022-09-27</a> using DeepL. If you looks something interesting but the auto-translated English is not good enough to understand it, feel free to let me know at <a href="https://twitter.com/nishio_en">@nishio_en</a>. I&#39;m very happy to spread my thought to non-Japanese readers.