ABCDEFGHIJKLMNOPQRSTUVWXYZ
1
2
3
4
5
과정 목표생성형AI는 어떻게 이미지를 만들까요?
6
7
8
MidjourneyDALL-E 3Stable DiffusionLLamVfirefly
9
10
11
1 시간Civitai:
https://civitai.com/
cyberrealistic
12
Hugging Face:
https://huggingface.co/
Multimodal->Text-to-Image->cyberrealistic
https://huggingface.co/cyberdelia/CyberRealistic
13
14
15
CLIP 에 대해 알아 보겠습니다.Cake Image
https://drive.google.com/file/d/1PItrv7ome4Hmr1mzemp8FnV4M7N4vgQm/view?usp=drive_link
16
Google Vision Api
https://cloud.google.com/vision?hl=ko
17
18
https://huggingface.co/stablediffusionapi
19
1 Models2개 모델 차이를 비교해 보세요
https://huggingface.co/stablediffusionapi/cyberrealistic
20
2
https://huggingface.co/stabilityai/stable-diffusion-xl-base-1.0
21
22
23
3https://getimg.ai/
24
Guides
https://getimg.ai/guides
25
Parameters
https://getimg.ai/guides/interactive-guide-to-stable-diffusion-guidance-scale-parameter
26
CFG
https://huggingface.co/spaces/google/sdxl
27
Seed
https://getimg.ai/guides/guide-to-seed-parameter-in-stable-diffusion
28
수박씨에서 Denoising 스텝 횟수 정의Steps
https://getimg.ai/guides/interactive-guide-to-stable-diffusion-steps-parameter
29
Negative prompts
https://getimg.ai/guides/guide-to-negative-prompts-in-stable-diffusion
30
31
32
https://civitai.com/images/1967197?modelVersionId=138176&prioritizedUserIds=6357&period=AllTime&sort=Most+Reactions&limit=20
33
영상 만들기 Prompt 연습
A handsome korean guy with short hair style.
34
gorgeous cute Austrian girl, smiling, (crop top), Steel gray hair loose braided hair, comics printed shirt, oversized jogger pants, lean against the wall, backstreet
35
36
37
Cyberrealistic image
https://civitai.com/images/1967197?modelVersionId=138176&prioritizedUserIds=6357&period=AllTime&sort=Most+Reactions&limit=20
38
39
AngrySad Laughing
40
41
SDXL image
https://civitai.com/images/2193181?modelVersionId=147872&prioritizedUserIds=227712&period=AllTime&sort=Most+Reactions&limit=20
42
43
44
45
눈 위치eye contact offront face
46
47
감성smilingangrylaughingsad
48
49
머리색Light Golden Blonde color hairsilver color hair
Chocolate Cherry color hai
50
51
옷with a tie and a white shirt
wearing a black jacket
52
53
나이an elegant middle aged manold manage 24
54
55
배경Meadowcityscape
56
57
목걸이안경
58
59
실습 이미지 생성
https://civitai.com/images/2712529
60
61
62
Huggin Face Spaces
63
64
5Guidance Scale을 바꾸어 보세요
https://huggingface.co/spaces/google/sdxl
65
66
6배경을 바꾸어 보세요
https://huggingface.co/spaces/adept/fuyu-8b-demo
67
Shopify 변경
https://ezfycode.com/blog/how-to-add-a-custom-background-image-to-any-shopify-theme
68
69
70
2 시간Prompts
https://docs.google.com/spreadsheets/d/14LMF7gTR0gnUCnpdWsCziMlhCCZuH_fisxruUM9oFdI/edit#gid=0
71
72
neck
https://drive.google.com/file/d/1pBaiZwAykXnPaA6sDQSz7AgYkxutnhmR/view?usp=sharing
73
74
3 시간Diffuser model의 이미지 생성 과정
75
76
DDPM과 모델학습DDPM
https://github.com/JSJeong-me/KOSA_Generate_AI_for_Image/blob/main/110_step_by_step.ipynb
77
78
모델학습
https://github.com/JSJeong-me/KOSA_Generate_AI_for_Image/blob/main/111_basic_training.ipynb
79
80
81
82
Diffusers 모델 활용Diffuser 모델활용
https://github.com/JSJeong-me/KOSA_Generate_AI_for_Image/blob/main/120-diffusers_intro.ipynb
83
84
Stable Diffusion - text-to-image
https://github.com/JSJeong-me/KOSA_Generate_AI_for_Image/blob/main/121-stable_diffusion.ipynb
85
runwayml/stable-diffusion-v1-5
86
Hugginface
stabilityai/stable-diffusion-2-1-base
https://github.com/JSJeong-me/Generate_AI_for_Image/blob/main/123-stable-diffusion-2-1-base.ipynb
87
88
stabilityai/stable-diffusion-2-1. This version can produce images with a resolution of 768x768, while the others work at 512x512.
89
90
91
4 시간Web UI 활용8Stable Diffusion - WebUI
https://github.com/JSJeong-me/Generate_AI_for_Image/blob/main/99-StableDiffusionWebUI.ipynb
92
93
https://df34d5038729551fa3.gradio.live
94
95
96
97
98
5 시간ControlNet7
https://huggingface.co/lllyasviel/sd-controlnet-canny
Girl with Pearl Earring
99
100
Text-to-Image Generation with ControlNet Conditioning