') + ')', 'gi'); if (regex.test(text)) { found = true; var frag = document.createDocumentFragment(); var parts = text.split(regex); parts.forEach(function(part, i) { if (i % 2 === 0) { frag.appendChild(document.createTextNode(part)); } else { var span = document.createElement('span'); span.className = 'userscript-highlight'; span.textContent = part; frag.appendChild(span); } }); node.parentNode.replaceChild(frag, node); } }); } else if (node.nodeType === 1 && node.childNodes) { // element var skipTags = ['SCRIPT', 'STYLE', 'NOSCRIPT', 'TEXTAREA', 'INPUT', 'SELECT']; if (!skipTags.includes(node.tagName)) { Array.from(node.childNodes).forEach(highlight); } } } highlight(document.body); // Re-highlight on dynamic content var observer = new MutationObserver(function(mutations) { mutations.forEach(function(m) { m.addedNodes.forEach(function(node) { if (node.nodeType === 1 || node.nodeType === 3) highlight(node); }); }); }); observer.observe(document.body, { childList: true, subtree: true }); })(); } } catch(__e) { console.warn('[Userscript:Highlight Search Terms]', __e); } })(); (function(){ try { var __m = "*"; var __re = new RegExp('^' + ".*" + ', 'i'); if (__m === '*' || __re.test(location.href)) { // Strip utm_, fbclid, gclid, etc. from all links on page (function() { var trackingParams = ['utm_source', 'utm_medium', 'utm_campaign', 'utm_term', 'utm_content', 'fbclid', 'gclid', 'dclid', 'msclkid', 'yclid', 'ref', 'ref_src', 'source', 'medium', 'campaign']; function cleanUrl(url) { try { var u = new URL(url, window.location.origin); var changed = false; trackingParams.forEach(function(p) { if (u.searchParams.has(p)) { u.searchParams.delete(p); changed = true; } }); return changed ? u.toString() : url; } catch (e) { return url; } } function cleanLinks() { document.querySelectorAll('a[href]').forEach(function(a) { var clean = cleanUrl(a.href); if (clean !== a.href) a.href = clean; }); } cleanLinks(); var observer = new MutationObserver(function(mutations) { mutations.forEach(function(m) { m.addedNodes.forEach(function(node) { if (node.nodeType === 1) { if (node.tagName === 'A') cleanLinks(); node.querySelectorAll('a[href]').forEach(function(a) { var clean = cleanUrl(a.href); if (clean !== a.href) a.href = clean; }); } }); }); }); observer.observe(document.body, { childList: true, subtree: true }); })(); } } catch(__e) { console.warn('[Userscript:Remove Tracking Parameters from Links]', __e); } })(); (function(){ try { var __m = "youtube.com"; var __re = new RegExp('^' + "youtube\\.com" + ', 'i'); if (__m === '*' || __re.test(location.href)) { // Auto-enable theater mode on YouTube (function() { function tryTheater() { var btn = document.querySelector('button[aria-label="Theater mode"], ytd-player #player button[title="Theater mode"]'); if (btn && !btn.classList.contains('activated')) { btn.click(); } } // Try immediately tryTheater(); // Try after navigation (SPA) var lastUrl = location.href; setInterval(function() { if (location.href !== lastUrl) { lastUrl = location.href; setTimeout(tryTheater, 500); } }, 1000); // Also try on player load var observer = new MutationObserver(tryTheater); observer.observe(document.body, { childList: true, subtree: true }); })(); } } catch(__e) { console.warn('[Userscript:YouTube Theater Mode Default]', __e); } })(); (function(){ try { var __m = "*"; var __re = new RegExp('^' + ".*" + ', 'i'); if (__m === '*' || __re.test(location.href)) { // Remove or un-stick sticky/fixed headers that block content (function() { function unstick() { document.querySelectorAll('header, nav, [role="banner"], .header, .navbar, .sticky, .fixed-top, [style*="position: fixed"], [style*="position:sticky"]').forEach(function(el) { if (el.style.position === 'fixed' || el.style.position === 'sticky' || getComputedStyle(el).position === 'fixed' || getComputedStyle(el).position === 'sticky') { el.style.position = 'static'; el.style.top = 'auto'; el.style.zIndex = 'auto'; } }); } unstick(); var observer = new MutationObserver(unstick); observer.observe(document.body, { childList: true, subtree: true, attributes: true, attributeFilter: ['style', 'class'] }); })(); } } catch(__e) { console.warn('[Userscript:Kill Sticky Headers]', __e); } })(); })(); GitHub - roberthoenig/dadata · GitHub
Skip to content
BranchesTags

Repository files navigation

TrueSkin: Accurately Visualizing Skin Conditions

Healthcare practitioners benefit from accurate skin disease modeling. Unfortunately, vanilla Stable Diffusion fails to suggest medically plausible visualizations.

TrueSkin addresses this problem by finetuning Stable Diffusion with textual inversion on manually collected datasets of skin conditions.

The finetuning takes minutes, the prediction seconds. The results outperform vanilla Stable Diffusion in medical realism.

Image 1Image 2Image 3Image 4
Base ImageActual Lupus Butterfly rashLupus Butterfly rash predicted by our ModelLupus Butterfly rash predicted by Stable Diffusion
Image 1Image 2Image 3Image 4
Base imageActual acne early stageAcne early stage prediction by our modelAcne early stage prediction by StableDiffusion

Implementation

Dataset

We manually collect data samples of the skin diseases Acne at different stages and Lupus. They are saved in Acne_progression and Lupus, respectively.

Finetuning

We want to accurately visualize skin conditions on people's faces.

To this end, we apply textual inversion with Stable Diffusion to finetune new text embeddings for skin conditions such as acne or Lupus.

We deploy https://github.com/AUTOMATIC1111/stable-diffusion-webui for the finetuning.

A detailed overview of textual inversion in stable-diffusion-webui is given here: https://github.com/AUTOMATIC1111/stable-diffusion-webui/wiki/Textual-Inversion.

Our model uses the following hyperparamters:

  • Embedding length: 2 tokens.
  • Embedding learning rate: 0.005.
  • Batch size: 2.
  • Prompt template: See prompt_template.txt.
  • Train for 500 - 4000 steps, stop when samples of sufficient quality are produced.

Web UI

Our WebUI is an adapted version of the one present in https://github.com/AUTOMATIC1111/stable-diffusion-webui. To install and run our code, follow the following steps:

  1. Follow the steps in https://github.com/AUTOMATIC1111/stable-diffusion-webui
  2. Copy the folder embeddings to stable-diffusion-webui
  3. Run webui.sh
  4. Add the script tampermonkey_js to your browser of choice, and click on the button Replace Textarea Content
  5. Generate new images!

About

No description, website, or topics provided.

Resources

Readme
Activity

Stars

4 stars

Watchers

1 watching

Forks

0 forks
Report repository

Releases

Packages

Contributors

Languages