尘白禁区-芬妮| Fenny |スノウブレイク：禁域降臨 | SNOWBREAK: CONTAINMENT ZONE | 塵白禁域|

가중치: 0.6-0.8, 권장값 0.7, 가중치가 높을수록 3D 감이 강해짐

권장 사용: 고해상도 복원(Hires. fix), controlnet-tile 이미지에서 이미지로 재생성

【강력히 추천하는 모델】: OnlyAnime |《唯》· 화려한 애니메이션 /model/105955/onlyanime-or

【부정적 임베딩】: UnrealisticDream, EasyNegative

CFG:7

vae: kl-f8-anime2

——————————————————————————————————————————————————————

【트리거 단어】

1. 긍정적 프롬프트: fenniS1CB, 1girl, blonde hair, medium hair, (ponytail:0.9), red eyes, dress, green ribbon, mini crown, bare shoulders, hairband, white gloves, detailed face,

선택적 긍정적 프롬프트: orange eyes, red arm bandage, singel red armband, thighhighs, high heel boots, elbow gloves, thigh strap, knee pads

2. 부정적 프롬프트: 3D, 3D rendering

선택적 부정적 프롬프트: weapon, holding, holding weapon, gun, holding gun, poor light

3. 부정적 임베딩: UnrealisticDream, EasyNegative

【수정 가능한 참조 프리셋】

1. 긍정적 프롬프트: fenniS1CB, 1girl, solo, tsundere, naughty face, standing, looking at viewer, sparkling eyes, red eyes, (blonde hair:1.1), medium hair, (ponytail:0.9), dress, green ribbon, white gloves, mini crown, hairband, elbow gloves, singel red armband, cowboy shot, masterpiece, best quality, (colorful), (finely detailed beautiful eyes and detailed face), cinematic lighting, extremely detailed CG unity 8k wallpaper, detailed face, lora:fenniSCB-v1:0.7

2. 부정적 프롬프트: (UnrealisticDream:1.2), EasyNegative, 3D, 3D rendering, weapon, holding weapon, gun, poor light

3. 기타 파라미터: DPM++SED Karras, 40 steps, 512x1024, Hires.fix 활성화, R-ESRGAN 4X+Anime6B, 1.2배 확대, 재생성 강도 0.35

——————————————————————————————————————————————————————

【현재 존재하는 문제 및 제안】

1. (1) 원거리 샷(전신)을 그릴 때, 인물의 눈과 얼굴 정확도가 때때로 결함이 생길 수 있으나, 아래 방법 【3】으로 보완 가능합니다.

(2) 인물의 머리 장식, 장갑, 하체 의상에 오류가 생길 수 있으나 전체적인 인물 느낌에 영향을 주지 않습니다.

(3) 프롬프트: red eyes는 orange eyes로 대체 가능합니다.

(4) sparkling eyes, tsundere, naughty face, smile 등의 프롬프트를 사용하여 인물의 표현력을 강화해 보세요.

2. 원거리 또는 전신 장면을 그릴 때, 이미지 크기와 스텝 수를 적절히 늘리면 인물의 얼굴 정확도 향상에 도움이 됩니다.

3. 위 방법 【2】의 효과가 만족스럽지 않을 경우, 생성된 이미지를 직접 이미지-이미지로 전송하여 재생성하십시오(효과 매우 우수), 방법은 다음과 같습니다:

(1) 원본 이미지와 동일한 프롬프트, 캐릭터 Lora, 시드를尽量 사용하세요.

(2) controlnet--tile 활성화, 확대 스크립트 Ultimate SD upscale 활성화

(3) 참조 설정 파라미터: Steps:30-40, 재생성 강도 0.25-0.4, 확대 알고리즘: 4x-UltraSharp 또는 R-ESGAN 4x+anime6B, 확대 배율 1.25--

4. NSFW 콘텐츠를 제작할 때, 실제 상황에 따라 가중치를 0.5-0.6로 낮추고, NSFW 관련 프롬프트를 더 많이 사용하며, NSFW 관련 프롬프트의 가중치를 높이고, dress, green ribbon, white gloves, mini crown, hairband, elbow gloves, singel red armband 등의 의상 관련 트리거 단어를 부정적 프롬프트에 추가하세요.

5. 이미지의 세부사항이 부족하거나 흐릿하다고 느낄 때, 반복 스텝 수를 적절히 증가시키거나, 샘플러 및 VAE를 조정하세요. 또는 방법 【3】을 이용해 이미지-이미지 재생성을 시도하세요.

6. Lora의 가중치가 낮을수록, 캐릭터 특징 관련 【트리거 단어】의 가중치를 높이는 것이 좋습니다. 예: blonde hair

7. 부정적 임베딩으로 UnrealisticDream, EasyNegative 사용을 강력히 권장합니다.

8. 모델의 일반성 문제 가능성

——————————————————————————————————————————————————————————————————————————————————————————————————————————————————

Weight: 0.6-0.8, 0.7 is recommended. The higher the weight, the stronger the 3D feeling.

Recommended uses: High-resolution fix (Hires. fix), controlnet-tile I2I

[Strongly recommended to use models]: OnlyAnime |《唯》· 炫彩动漫 /model/105955/onlyanime-or

【Negative embaddings】: UnrealisticDream, EasyNegative

CFG:7

vae: kl-f8-anime2

___________________________________________________________________________________________________________________________________________________________________

【Trigger Words】

1. Positive prompts: fenniS1CB, 1girl, blonde hair, medium hair, (ponytail:0.9), red eyes, dress, green ribbon, mini crown, bare shoulders,hairband,white gloves,detailed face,

Optional prompt words on the front: orange eyes, red arm bandage, singel red armband, thighhighs, high heel boots, elbow gloves, thigh strap, knee pads

2. Negative prompt words: 3D, 3D rendering

Negative optional prompt words: weapon, holding, holding weapon, gun, holding gun, poor light

3. Negative embaddings: UnrealisticDream, EasyNegative,

______________________________________________________________________________________________

[A reference preset that you can modify on top of]

1. Positive prompt words: fenniS1CB, 1girl, solo, tsundere, naughty face, standing, looking at viewer, sparkling eyes, red eyes, (blonde hair:1.1), medium hair, (ponytail:0.9), dress, green ribbon, white gloves, mini crown, hairband, elbow gloves, singel red armband, cowboy shot, masterpiece, best quality, (colorful),(finely detailed beautiful eyes and detailed face),cinematic lighting, extremely detailed CG unity 8k wallpaper , detailed face, lora:fenniSCB-v1:0.7

2. Negative prompt words: (UnrealisticDream:1.2), EasyNegative, 3D, 3D rendering, weapon, holding weapon, gun, poor light

3. Other parameters: DPM++SED Karras, 40 steps, 512x1024, enable Hires.fix, R-ESRGAN 4X+Anime6B, zoom in 1.2 times, redraw range 0.35

————————————————————————————————————————————————————————————————————————————————

[Current problems and suggestions]

1. (1) When drawing a picture of a long-range shot (full body), there may be defects in the accuracy of the eyes and faces of the characters, but they can be repaired in the following way [3].

(2) There may be errors in the costumes of the character's headgear, gloves and legs, but it will not affect the overall character look and feel.

(3) Prompt word: red eyes can be replaced by orange eyes.

(4) You can try to use prompt words such as sparkling eyes, tsundere, naughty face, smile, etc. to enhance the perception of characters.

2. When drawing a distant scene or a full-body picture, please increase the size and steps of the picture appropriately, which will help improve the accuracy of the character's face.

3. If the effect of the above method [2] is not satisfactory, please directly send the generated picture to "image to image" for redrawing [the effect is very good], the method is as follows:

(1) Try to use the prompt words, character lora, and seed that are consistent with the original picture

(2) Enable controlnet--tile, enable the zoom script Ultimate SD upscale

(3) A reference setting parameter: Steps:30-40, redrawing range 0.25-0.4, magnification algorithm: 4x-UltraSharp or R-ESGAN 4x+anime6B, magnification factor 1.25--

4. When creating NSFW content, please reduce the weight to 0.5-0.6 according to the actual situation, and use more NSFW-related prompt words at the same time, increase the weight of NSFW-related prompt words, and set dress, green ribbon, white gloves, mini crown, Hairband, elbow gloves, singer red armband and other clothing trigger words are added with negative prompts.

5. When you feel that the details of the picture are insufficient or blurry, please increase the number of iteration steps appropriately, or adjust the sampler and vae, or try the method [3] to redraw the picture.

6. The lower the weight of the Lora, the more the weight of the character feature [Trigger Words] should be increased, for example: blonde hair

7. It is strongly recommended to use negative embaddings: UnrealisticDream, EasyNegative

8. Possible problems with model versatility