shisei
π Free CC0 mannequin pose reference library for AI image tools (Nano Banana, ChatGPT, ComfyUI, Midjourney) β 250+ poses Γ 5 angles in line art, names in ja/zh/en/ko
- ai-art
- ai-image-generation
- art-reference
- cc0
- comfyui
- controlnet
- dataset
- drawing-reference
- figure-drawing
- line-art
- mannequin
- midjourney
- multilingual
- nano-banana
- pose-reference
- reference-images
- stable-diffusion
- Stars
- 400
- Forks
- 83
- + today
- +4
- Created
- 6d
Ranking data as of November 3, 2026 (UTC).
Star History
Today, hour by hour
Overview
π Free CC0 mannequin pose reference library for AI image tools (Nano Banana, ChatGPT, ComfyUI, Midjourney) β 250+ poses Γ 5 angles in line art, names in ja/zh/en/ko It ranks #80 on GitTiger, gaining +4 stars on November 3, 2026 (UTC).
The project is written in Python and has 83 forks. It was created 6d ago.
What it is
Shisei (ε§Ώε’) is a free mannequin pose reference library for AI image tools. Its description names Nano Banana, ChatGPT, ComfyUI and Midjourney, and cites 250+ poses at 5 angles each as line art, with names in Japanese, Chinese, English and Korean. The README summary states 280 poses in four series, each with five camera angles, drawn as clean line art.
Capabilities described
- **Line-art pose views:** five camera angles per pose. - **Quality review:** every view is reviewed (docs/qc/); failing views are excluded and left empty rather than filled with a bad drawing. - **Redraws:** 43 sets were redrawn by a second generator. - **Photo renders:** 89 renders for std P001βP022 and P050, described as AI-generated adults; P019, P020 and P022 are off because they show the old drawing.
Tags include ControlNet, ComfyUI, figure drawing and dataset. The description calls it CC0, while the licence metadata lists Apache-2.0.
Getting started
No setup steps were supplied. See the repository README for usage.
git clone https://github.com/wquguru/shisei.git
cd shisei
# see README for setupFrequently Asked Questions
What is Shisei?
A free mannequin pose reference library for AI image tools, offering line-art poses at five camera angles, per its description and README summary.
How many poses does it contain?
The description says 250+ poses Γ 5 angles, and the README summary says 280 poses in four series Γ 5 camera angles.
What licence applies?
The description calls the library free CC0, but the licence metadata lists Apache-2.0. The supplied data does not reconcile them, so check the repository.
Which languages are the pose names in?
The description says names are in Japanese, Chinese, English and Korean.