Seeking Reliable Character Swap & Pose Preservation Workflow
MiRAGamingChannel · reddit · 2026-07-14
The author is looking for a reliable ComfyUI image-to-image workflow that achieves "character replacement while strictly preserving the original pose and background," sharing recent testing frustrations.
Tested solutions and pain points:
- SDXL pipeline (with GroundingDINO, SAM, ControlNet, and InstantID): Runs but shows significant ghosting and weak character similarity.
- Flux.2 Klein 9B (community workflow): Identity and clothing transfer are decent, but textures look flat and plastic at low step counts.
- Qwen-Image-Edit-2511 (three-stage pipeline): Best texture quality, but regardless of prompt adjustments, the output is always dominated by the "donor" image's pose and background, failing to strictly preserve the target image's pose.
Running on 16GB VRAM and limited by compute, the author hopes to find a truly reliable solution that doesn't require constant parameter tuning.
More from Multimodal
- HeyGen adds a media-sourcing skill for coding agents with 75k images and 10k tracks — HeyGen · 2026-07-22
- Reddit user seeks ComfyUI NSFW text-to-image and image-to-video workflows under 20 GB VRAM — hobbyist2020 · 2026-07-22
- Krea 2 users recommend a two-pass Clownshark sampler setup for sharper image details — listopalafoto · 2026-07-22
- Gemini Omni Flash turns a boat cabin into a cave in Flow by Google — chrisfirst · 2026-07-22
- A simple workflow to turn a photo into an image prompt using Gemini, Grok, or GPT Image — harshitagu72595 · 2026-07-22
- A Reddit user proposes a consistency LoRA to keep anime and game scenes visually stable — ThirdWorldBoy21 · 2026-07-22