1，本模型所有数据全来视频截图

2，最大出图大小建议不要超过 1280 x 1280

3，建议比重 在0.7-0.9左右脸和衣服还原不够提高比重，

4，采样方式 DPM＋＋2Mkarras   步数  20-30  重绘-0.3-0.5

5，算法 4×-UltraShrt   高分步数 15-20  

展示图描述词

(1girl:1.1),(looking at viewer:1.4),solo,(hair ornament:1.4),

(white hair:1.4),(long hair:1.6),(necklace:1.6),(hair_bun:1.4),(jewelry:1.6),(crystal:1.4),(water_eyes:1.2),(brown eyes:1.2),(facial mark:1.4),(forehead mark:1.4),(earrings:1.4),(dress:1.6),(detached sleeves:1.6),(purple dress:1.4),(white dress:1.4),(bare shoulders:1.4),(long_skirt:1.4),(fluffy:1.3),(frilled:1.4),hair_ribbon,(spandex:1.2),(waterfall:1.5),(castle:1.1),(cityscape:1.6),(sky:1.4),(white clouds:1.6),(cherry_blossoms:1.3),(sakura tree:1.6),(river:1.4),(flowers and trees:1.3),(petals:1.5),(water:1.6),