"스테이블 디퓨전, AnimateDiff 사용 연습 1)"

스테이블 디퓨전과 AnimateDiff로 변하는 이미지를 만들어 보았습니다. 아래는 제가 적용해본 비율과 프롬프트들입니다. 확인해보세요.

"스테이블 디퓨전, AnimateDiff 사용 연습 1)"

; a very cool car <lora:lcm-lora-sd1-5:1>

a very cool car <lora:lcm-lora-sd1-5:1>

0: red collar
8: blue collar
16: black collar
24: brown collar
Steps: 8, Sampler: DPM++ 2M, Schedule type: Karras, CFG scale: 7, Seed: 2263964953, Size: 768x512, Model hash: f47e942ad4, Model: realisticVisionV60B1_v51HyperVAE, VAE hash: 8495cebb2c, VAE: vaeFtMse840000EmaPruned_vae.safetensors, Denoising strength: 0.75, Clip skip: 2, AnimateDiff: "model: mm_sd_v15_v2.ckpt, video_length: 32, fps: 8, loop_number: 0, closed_loop: R+P, batch_size: 16, stride: 1, overlap: 4, interp: Off, interp_x: 10, freeinit_enable: False, mm_hash: 69ed0f5f, latent_power: 1, latent_scale: 32, latent_power_last: 1, latent_scale_last: 32, version: v2.0.0.2-a-4-g69a2395", Mask blur: 58, Lora hashes: "lcm-lora-sd1-5: aaebf6360f7d", Version: v1.9.4

스테이블 디퓨전 ; Mask blur 58

스테이블 디퓨전 ; Mask blur 8

스테이블 디퓨전 ; Mask blur 1

 

 

==

masterpeice, high quility, uhd, 8k,1girl,

0: T-shirt
8:
16: T-shirt
24:
Negative prompt: worst quility, low quility, nsfw, Steps: 25, Sampler: DPM++ 2M, Schedule type: Karras, CFG scale: 7, Seed: 3070722133, Size: 512x512, Model hash: f47e942ad4, Model: realisticVisionV60B1_v51HyperVAE, VAE hash: 8495cebb2c, VAE: vaeFtMse840000EmaPruned_vae.safetensors, Denoising strength: 0.75, Clip skip: 2, AnimateDiff: "model: mm_sd_v15_v2.ckpt, video_length: 40, fps: 8, loop_number: 0, closed_loop: R+P, batch_size: 16, stride: 1, overlap: 4, interp: Off, interp_x: 10, freeinit_enable: False, mm_hash: 69ed0f5f, latent_power: 1, latent_scale: 32, latent_power_last: 1, latent_scale_last: 32, version: v2.0.0.2-a-4-g69a2395", Mask blur: 49, Version: v1.9.4

 

==

masterpeice, high quility, uhd, 8k,1girl, 0: T-shirt 8: T-shirt 16: Tank top 24: Tank top Negative prompt: worst quility, low quility, nsfw, Steps: 25, Sampler: DPM++ 2M, Schedule type: Karras, CFG scale: 7, Seed: 3070722133, Size: 512x512, Model hash: f47e942ad4, Model: realisticVisionV60B1_v51HyperVAE, VAE hash: 8495cebb2c, VAE: vaeFtMse840000EmaPruned_vae.safetensors, Denoising strength: 0.75, Clip skip: 2, AnimateDiff: "model: mm_sd_v15_v2.ckpt, video_length: 32, fps: 8, loop_number: 0, closed_loop: R+P, batch_size: 16, stride: 1, overlap: 4, interp: Off, interp_x: 10, freeinit_enable: False, mm_hash: 69ed0f5f, latent_power: 1, latent_scale: 32, latent_power_last: 1, latent_scale_last: 32, version: v2.0.0.2-a-4-g69a2395", Mask blur: 8, Version: v1.9.4

 

"스테이블 디퓨전, AnimateDiff 사용 연습 2) ; 바다 속에서 수영"

"스테이블 디퓨전, AnimateDiff 사용 연습 2) ; 바다 속에서 수영"

 

스테이블 디퓨전과 AnimateDiff를 활용한 변하는 이미지 생성 과정을 통해 AI 기술의 무한한 가능성을 확인할 수 있었습니다. 이 글이 여러분의 창작 프로젝트에 영감을 주길 바라며, 다양한 시도를 통해 더욱 창의적인 결과물을 만들어 보시기 바랍니다.

 

 

"스테이블 디퓨전과 AnimateDiff로 창의적인 얼굴 표정과 옷 변화 영상 만들기"

"스테이블 디퓨전과 AnimateDiff로 창의적인 얼굴 표정과 옷 변화 영상 만들기"

 

"Stable Diffusion의 인페인팅과 아웃페인팅: 이미지 편집의 혁신적인 기술"

"Stable Diffusion의 인페인팅과 아웃페인팅: 이미지 편집의 혁신적인 기술"