ComfyUI trick: aux preprocessor + Qwen transfers poses across characters with one prompt
Acceptable-Work8202 · reddit · 2026-09-23
A practical ComfyUI combo the author says isn't widely known: load a pose reference (e.g. a dwpose or depthmap output) into a default aux preprocessor node, wire it into Qwen's image1 input, put the character you want drawn into image2, and prompt something like "Draw the character from image2, use the pose of image1."
The character in image2 gets redrawn in image1's pose. The prompt can be amended as needed for other edits.
More from Multimodal
- Given a 3-hour budget and a one-line prompt, Opus 5.5 produced a full Kowloon horror short itself — rainbird · 2026-09-23
- Image-gen prompt sharer techhalla posts 'safer highways' generations with prompts — techhalla · 2026-09-23
- Opus 5.5 recreates early-2000s stick figure fight YouTube animations — JasonBotterill · 2026-09-23
- 'Ordinary': An AI-generated Kowloon-set horror short turns heads on Reddit — rainbird · 2026-09-23
- Screenshot any character, one prompt recreates it — iamaliveix · 2026-09-23
- Google open-sources Noto 3D emoji — sculpting revealed their poodle only ever had three legs — GoogleDesign · 2026-09-23