This workflow converts any non-realistic images to realistic using Pony XL.<img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/d9995f24-96d6-45d0-84c7-aaeab6ef004f/width=525/d9995f24-96d6-45d0-84c7-aaeab6ef004f.jpeg" /><h1 id="what's-new-in-v3">🆕 What's New in V3</h1>Welcome to Pony-Anime-Cosplay-v3 workflow! Below are the key improvements and updates designed to optimize the whole workflow, improve facial recognition, and more features.<hr /><h2 id="refiner">📈 Refiner</h2><ul><li>Added SDXL Refiner for more realism regardless of mode - uses JuggernautXL</li><li>Utilizes InsightFace and FaceID PLUS V2 for facial detection and face swap.</li><li>Uses Tile or Canny CNet for more accuracy - switch between two depending on the refiner output - defaults to Tile.</li><li>Switch between initial generated face (from Pony) or an image input.</li></ul>⚠️ Note: Face detection may occasionally fail or detect the wrong face. If results are inconsistent, try different inputs (main and/or face image).<hr /><h2 id="film-grain-integration">🌾 Film Grain Integration</h2><ul><li>Uses VRGameDevGirl Fast Film Grain (set to 0.02) to add more realistic output.</li></ul><hr /><h2 id="configuration-changes">⚙️ Configuration Changes</h2>💡 These are based on personal preferences. Feel free to adjust them to fit your workflow needs.<ol><li>⏭️ Moved Highres-Fix to Refiner - it doesn't make sense to put on the first generation.</li><li>❌ Disabled cinematic photography... Lora.</li><li>➕ Added separate seed node to each detailer for more flexibility. I use it for testing purposes but can be useful if you want to regenerate a certain detailer.</li><li>🔄 Changed values for each CNet nodes.</li><li>➕ Added Additional Promp which appends to the end of final prompt. This is already available in the text concat, but due to the node placements, i just added it as a connection to the 6th concat. This allows us to be more precise, for example, character names if it wasn't added, or other charactre features missed by the tagger.</li></ol><hr /><h2 id="quality-of-life-improvements">🧰 Quality-of-Life Improvements</h2><ol><li>🧩 General node rearrangement for a cleaner layout.</li><li>🔗 Updated Modes &amp; Download Links – including updated default settings.</li><li>➕ Added new modes!</li></ol><h1 id="always-start-with-the-usage-info-inside-the-workflow.">📖 Always start with the USAGE info inside the workflow.</h1><div data-youtube-video><iframe width="640" height="480" allowfullscreen="true" autoplay="false" disablekbcontrols="false" enableiframeapi="false" endtime="0" ivloadpolicy="0" loop="false" modestbranding="false" origin playlist src="https://www.youtube.com/embed/DgT3PBFObSY" start="0"></iframe></div>

Cosplay-Workflow-Pony

<pre><code># 🆕 What's New in V3

Welcome to Pony-Anime-Cosplay-v3 workflow! Below are the key improvements and updates designed to optimize the whole workflow, improve facial recognition, and more features.

---

## 📈 Refiner
- Added SDXL Refiner for more realism regardless of mode - uses **JuggnernautXL**
- Utilizes **InsightFace** and **FaceID PLUS V2** for facial detection and face swap.
- Uses Tile or Canny CNet for more accuracy - switch between two depending on the refiner output - defaults to Tile.
- Switch between initial generated face (from Pony) or an image input.

⚠️ *Note: Face detection may occasionally fail or detect the wrong face. If results are inconsistent, try different inputs (main and/or face image).*

---

## 🌾 Film Grain Integration
- Uses **VRGameDevGirl** Fast Film Grain (set to 0.02) to add more realistic output.

---

## ⚙️ Configuration Changes 
&gt; 💡 *These are based on personal preferences. Feel free to adjust them to fit your workflow needs.*

1. ⏭️ Moved *Highres-Fix* to Refiner - it doesn't make sense to put on the first generation.
1. ❌ Disabled *cinematic photography...* Lora.
1. ➕ Added separate seed node to each detailer for more flexibility. I use it for testing purposes but can be useful if you want to regenerate a certain detailer.
1. 🔄 Changed values for each CNet nodes.
1. ➕ Added **Additional Promp** which appends to the end of final prompt. This is already available in the text concat, but due to the node placements, i just added it as a connection to the *6th* concat. This allows us to be more precise, for example, character names if it wasn't added, or other charactre features missed by the tagger.

---

## 🧰 Quality-of-Life Improvements

1. 🧩 General **node rearrangement** for a cleaner layout.
1. 🔗 Updated **Modes &amp; Download Links** – including updated default settings.
1. ➕ Added new modes!

# 📖 Always start with the USAGE info inside the workflow.
# ▶️ Video Tutorial to follow.</code></pre>

Files with SHA256: f065f8cd1b3f4fe9499c1d4b86add1da8a662113b1211e303eb01f68c35dc27e