일괄성있는 케릭터 그리기
기본 프롬프트는 아래와 같지만 베이스 모델과 로라모델을 적절히 잘 이용해야 원하는 그림을 얻을 수 있습니다.
<lora:Ink scenery:0.55>, <lora:add_detail:1.51>, <lora:film:0.79>, <lora:add_detail:0.49>, <lora:film:0.59>, <lora:add_detail:0.09>, (photorealistic:1.4), (best quality:1.3), (extremely intricate:1.2), (ultra highres:1.2), ((korean beauty model:1.6)), (small head), (almond-shaped eyes), (symmetric eyes, complex pupils, extremely detailed eyes and face, light on face, sharp nose, nose blush), ((korean idol:1.4)), [:(detailed face:1.2):0.2], (sharp focus:1.4), (huge breast), (wide pelvic:1.3), (long legs), ((((pretty korean girl)))), (((looking at viewer))), (((((random angle))))), BREAK\n ((((denim, jeans, 1girl, solo, pants, high_heels, long_hair, sitting, black_hair, torn_clothes, torn_jeans, traditional_media, hand_in_hair, closed_eyes, hand_on_own_head, artist_name, photo, full_body, shoes, watermark, lips, torn_pants, realistic, crossed_legs, yellow_shirt, long_sleeves, adjusting_hair, shirt, no_socks, nose, casual, white_footwear, marker_(medium), smoke, high_heel_boots, orange_shirt, blue_pants, bangs, breasts, signature, sweater, indoors, medium_breasts, copyright_name, closed_mouth, web_address, english_text, chair, oldschool, building, dated, stiletto_heels, window, hand_up, parted_lips, alternate_costume, socks, shadow, against_wall, collarbone, pink_lips, knee_up, smile, hair_over_one_eye, head_rest, arm_up, leg_up, arm_at_side, asian, character_name, shoe_dangle, lipstick, boots, hand_behind_head, makeup, profile, brown_hair, looking_at_viewer, cigarette))))
Negative prompt: EasyNegative, EasyNegativeV2, FastNegativeV2, ng_deepnegative_v1_75t, badhandv4, paintings, sketches, ((worst quality:2), (low quality:2), (normal quality:2), lowres, normal quality, (monochrome), (grayscale), skin spots, acnes, skin blemishes, age spot, manboobs, double navel, muted arms, fused arms, analog, analog effects, extra arms, extra hands, extra legs, extra foots, bad architecture, watermark,(mole:1.5), ((blurry)),
Steps: 55, Sampler: Euler a, CFG scale: 6.5, Seed: 709338829, Size: 576x1024, Model hash: 5b53949fc6, Model: astranime_V6, VAE hash: b8821a5d58, VAE: vae-ft-mse-840000-ema-pruned.safetensors, Denoising strength: 0.4, Clip skip: 2, DDetailer model a: bbox\mmdet_anime-face_yolov3.pth [51e1af4a], DDetailer conf a: 30, DDetailer dilation a: 4, DDetailer offset x a: 0, DDetailer offset y a: 0, DDetailer preprocess b: False, DDetailer bitwise: None, DDetailer model b: segm\mmdet_dd-person_mask2former.pth [3848e764], DDetailer conf b: 30, DDetailer dilation b: 4, DDetailer offset x b: 0, DDetailer offset y b: 0, DDetailer mask blur: 4, DDetailer denoising: 0.4, DDetailer inpaint full: True, DDetailer inpaint padding: 32, DDetailer cfg: 7, Script: Detection Detailer, Lora hashes: "makina69_katokonatsu_v1.0: ae96b2e96754, add_detail: 7c6bad76eb54, add_detail: 7c6bad76eb54, add_detail: 7c6bad76eb54", TI hashes: "easynegative: c74b4e810b03, EasyNegativeV2: 339cc9210f70, FastNegativeV2: a7465e7cc2a2, ng_deepnegative_v1_75t: 54e7e4826d53, badhandv4: 5e40d722fc3d", Downcast alphas_cumprod: True, Mask blur: 4, Inpaint area: Only masked, Masked area padding: 32, Version: v1.8.0