At the start I'm training only pixelated censoring. I would like to first fine tune dataset/training on understanding of the concept of censoring on any object/part of object with natural language.<h2 id="version-comparison-fwjal6z3c">Version comparison</h2><ul><li>ALPHA2 is much more flexible than ALPHA1</li><li>ALPHA2 is able to censore things that were not present in training dataset</li><li>ALPHA2 has a bit of issues with hands and quality (jpeg artifacts)</li><li>ALPHA2 is not censoring things that you do not specify as ALPHA1 did</li><li>ALPHA2 does not censore somtimes entire objects (mostly on bigger ones)</li></ul><h2 id="how-to-use-nd10gia8f">How to use</h2><ul><li>Trigger word censored</li><li>It works pretty well with natural language prompts (prompts that seemed to work for me are adding to the end "&lt;something&gt; is/are censored.", before the thing you wanna censore add "censored &lt;something&gt;", or add right behind "&lt;somthing&gt; that is /are censored")</li><li>Don't use this LoRA on clip or keep it on strength 1</li></ul><h2 id="known-issues-5o1wd3m5w">Known issues</h2><ul><li>Jpeg artifacts in certain parts of the image</li><li>Hands are generated blurry or with wrong number of fingers</li></ul>If you have some ideas for future types of censoring leave them in the comments.Enjoy :)

Files with SHA256: 436a82b8ffc3d86bcb1217543040d501a8deb5eee2b7d637778d3496b253a551