<img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/9236d00b-cd55-4700-a0dc-bc0ddc9aa567/width=525/9236d00b-cd55-4700-a0dc-bc0ddc9aa567.jpeg" /><span style="background-color:rgb(26, 27, 30);color:rgb(193, 194, 197);font-family:-apple-system, BlinkMacSystemFont, &quot;Segoe UI&quot;, Roboto, Helvetica, Arial, sans-serif, &quot;Apple Color Emoji&quot;, &quot;Segoe UI Emoji&quot;;font-size:16px">Using the power of Qwen Edit, inject a character seamlessly into a scene, properly adapting to the scene's lighting conditions! The goal is the find a good way to crop a character, inject it into a scene, then relight the character and generate the shadows and/or reflections if need be. I've made this workflow as a 'second step' to my <a target="_blank" rel="ugc" href="https://civarchive.com/models/1982115?modelVersionId=2325436">repose workflow</a>, but you can use any character image you like.You can position the character wherever you want, at whatever scale you want in the scene. Its designed so that you can place your character anywhere on a massive background, which you can make several of, then crop out the 'shots' you need as initial images for whatever vid gen model you prefer.I'm only gonna put basic info here, for details, there are notes within the workflow.With this one, you gotta embrace the gotcha, because shadows be random.Tutorial:= coming soon =<h3 id="version-1.0">VERSION 1.0</h3><ul><li>The workflow uses the Qwen Edit 2509 All-in-One model by Phr00t, found <a target="_blank" rel="ugc" href="https://huggingface.co/Phr00t/Qwen-Image-Edit-Rapid-AIO/tree/main/v5">here</a>.</li><li>Upscale model you can use whatever you like. I use <a target="_blank" rel="ugc" href="https://huggingface.co/FacehugmanIII/4x_foolhardy_Remacri/tree/main">Recremari x4</a>.</li></ul>Inputs:<ul><li>Character Image<ul><li>Best if fully visible on white background</li></ul></li><li>Background Image<ul><li>Can be as big as you need, but the foreground shouldn't obscure the character</li></ul></li><li>Lighting, Shadow and Reflection Prompt<ul><li>You need to revise the prompt to describe the lighting, tint, shadow, and reflections. There is a note inside to guide you on how.</li></ul></li></ul>Outputs:<ul><li>Character on Background</li><li>Cropped Character on Background</li><li>Results Info Sheet</li></ul><h3 id="require-resources:">Require Resources:</h3>Version 3.0📂 ComfyUI/└── 📂 models/ └── 📂 checkpoints/ |── Qwen-Rapid-AIO-NSFW-v5.safetensors | or └── Qwen-Rapid-AIO-NSFW-v5.safetensorsStay Cheesy Y'all!~<ul><li>Brie Wensleydale</li></ul>

Brie's Qwen Edit Lazy Relight

Initial VersionUsing Qwen Edit 2509 All-In-One modelUse whatever upscaler you like, I use recrimari 4x

Files with SHA256: d8b51cf55240ebcd52eb5072c54e5c6fdd87f2d83136b03fd6dbad42831e3f39