사용자 LoRA

XL
디테일 강화

Anime-Hands- old, use improved

#사이버 패션
Anime-Hands- old, use improved - AI Model 커버 이미지
업로드 날짜
2025. 7. 2. 오후 10:23
이용수
124
획득
10,117
리뷰
리뷰 없음
트리거 단어

hands, anime_hands, fist, V-sign, Rocker-Hands

권한
  • 이미지 생성 및 공유 허용
  • 다른 사용자가 내 모델을 다운로드하도록 허용합니다
  • 생성된 이미지의 상업적 이용 허용

설명

This is my test experiment at a hand-detailer improvement model. If it actually works, I might make more or at least a more extensive one.


Note: While it does actually work, I made improved version of the model 'anime hands-improved' preference and update will be given to that one.


Testing out the experiment by doing hands only as the image input on transparent background. As such, you will need base model, character loras, style loras and other negative/enhancers as well.


Trained on 30 images, 10 of each pose, tried doing 3D-360 view of each pose. Most shots covered from finger tip to elbow, all included the wrist and upper forearm. VRoid is not incredibly high poly or detailed on model I used with texture, but the outlines and ambient occlusion of posing app should give it the needed discerning details.


Trained on SDXL with Haruka v2, recommended stats are unknown as of yet due to experimental nature. Tags and triggers associate to the 'pose' of the hand. Model is trained on 3 poses.


  1. V-sign, aka 'peace sign' or 'victory hands'
  2. standard fist
  3. the rocker sign aka 'devil horns' hand sign.

댓글

Anime-Hands- old, use improved으로 만든 작품