Comic Avatar Design - Half Body Headshot

[Model introduction]

The LoRa training set consists of various comic styles with big head effects. Early comic avatars with exaggerated effects of real characters are calculated. Combining large portrait-like models can stably generate realistic, hand-drawn, thick paint, or 2.5D style bust portraits.

The model generalization test is comprehensive, and hair color, clothing, accessories, eye color, etc. can be freely replaced; [Model drawing parameters] Large model: AWPaintbarley realistic/chilloutMix, etc. (portrait and painting are all possible) CLIP skip layer: 1-2 VAE: Select sampling method as needed: Eulera number of iteration steps: 18-22 steps Recommended plotting size: 806*1120 (different large models (Multiple limbs or multiple heads may appear below, which can be solved by adjusting the resolution size) Prompt guidance coefficient: 7.0 [plot keyword] trigger word: datou positive hint Words (reference): 1 boy, male focus, solo, necktie, formal, belt, shirt, blue necktie, smile, suit, realistic, jacket, pants, brown hair, looking at viewer, white shirt, vest, black jacket, simple background, hands in pockets, collared shirt, black pants, brown eyes, teeth, grin, short hair, black suit, long sleeves, grey background, facial hair, datou, (best quality: 1.5), (masterpiece: 1.5), (super realistic: 1.5), (high detail: 1.5) negative hint words: (worst quality, low quality: 2), (normal quality: 1.5), monochrome, (watermark: 1.4)), artist name, character name, text, bad anatomy, sketch, duplicate, (ng_deepnegative_v1_75t), negative_hand-neg, bad_pictures, easyNegative,