Seeking ComfyUI Workflows for True Photorealistic Style Transfer from 3D CGI
Ellokko87 · reddit · 2026-11-17
A Reddit user asks for ComfyUI workflows or tips to transform low-detail 3D CGI comics into convincing photorealistic images—true stylistic reinterpretation rather than filtering, upscaling, or enhancing. The transformation should preserve characters, identities, poses, composition, objects, and text while completely reimagining the visual style, and ideally generalize across scenes without manual per-image prompting.
They've tried Qwen Image Edit and FLUX, but results retain too much of the original CGI look; they note the new ChatGPT does an excellent job here and wonder if any model, specialized LoRA, or clever ComfyUI workflow can keep up.
More from Multimodal
- REA-Video skill lets Claude reverse-engineer video editing styles; rival claims 20x speed at SSIM 0.96 — toolstelegraph · 2026-11-17
- Songwriter finishes AI music video from hospital bed with Suno and Runway — Kyrannio · 2026-11-17
- Ant Group open-sources Duo, a reciprocal world model where virtual agents decide when to speak — aigclink · 2026-11-17
- One sentence in, a voiced 30-second animated short out: the 5-part storyboard prompt worth stealing — MichelleZhao91 · 2026-11-17
- OpenMotion open-sourced: 470K lines in 14 days, claims 95% replication of viral video motion — yihui_indie · 2026-11-17
- AI short film on medieval rushlight lighting, made entirely locally — MandelDuck · 2026-11-17