Sign In

ShinYunBok

225
1.9k
11
Verified:
SafeTensor
Type
Checkpoint Trained
Stats
184
Reviews
Published
Jun 2, 2023
Base Model
SD 1.5
Training
Steps: 1,918
Epochs: 7
Trigger Words
shinyunbok painting
Hash
AutoV2
3DC0F91D98
default creator card background decoration
bebechien's Avatar
bebechien

Shin Yun-bok, better known by his pen name Hyewon (1758–1813), was a Korean painter of the Joseon Dynasty. Like his contemporaries Danwon and Geungjae, he is known for his realistic depictions of daily life in his time.

Suggested prompt: "best quality, shinyunbok painting, <your prompt>"
Suggested negative prompt: "worst quality, low quality"
You can adjust the strength, like (shinyunbok painting:0.7) to weaken the style.

v1.0, v2.0 - Train Base model : https://huggingface.co/runwayml/stable-diffusion-v1-5
v2.0 Animation - Train Base model : SilvermoonMix Base v1.2
v2.0 Animation SD2.1 - Train Base model : Replicant-V3.0
SDXL - Train Base model : https://huggingface.co/stabilityai/stable-diffusion-xl-base-1.0

========================================================================

For v2.1 specific - Train Base model : Waifu Diffusion - Beta 03

Suggested prompts: "shinyunbok anime, (exceptional, best aesthetic, new, newest, best quality, masterpiece, extremely detailed, waifu:1.2), <your prompt>"
Suggested negative prompt: "((photo, realistic)), lowres, ((bad anatomy)), ((bad hands)), missing finger, extra digits, fewer digits, blurry, ((mutated hands and fingers)), (poorly drawn face), ((mutation)), ((deformed face)), (ugly), ((bad proportions)), ((extra limbs)), extra face, (double head), (extra head), ((extra feet)), monster, logo, cropped, worst quality, jpeg, humpbacked, long body, long neck, ((jpeg artifacts)), deleted, old, oldest, ((censored)), ((bad aesthetic)), (mosaic censoring, bar censor, blur censor)"

v2.1 mercury SD2.1 - strong Shin Yun-bok style
v2.1 venus SD2.1 - slightly weak Shin Yun-bok style and leaning on Japanese anime style

========================================================================

All trained images are from https://gongu.copyright.or.kr/
Training Datasets : https://huggingface.co/datasets?search=bebechien