kyujinpy's picture
Upload README.md
691e4f8
metadata
license: creativeml-openrail-m
base_model: kyujinpy/KO-anything-v4-5
training_prompt: A bear is playing guitar
tags:
  - tune-a-video
  - text-to-video
  - diffusers
  - korean
inference: false

Tune-A-VideKO-anything

Github: Kyujinpy/Tune-A-VideKO

Model Description

Samples

sample-500 Test prompt: 1์†Œ๋…€๋Š” ๊ธฐํƒ€๋ฅผ ์—ฐ์ฃผํ•˜๊ณ  ์žˆ๋‹ค, ํฐ ๋จธ๋ฆฌ, ์ค‘๊ฐ„ ๋จธ๋ฆฌ, ๊ณ ์–‘์ด ๊ท€, ๊ท€์—ฌ์šด, ์Šค์นดํ”„, ์žฌํ‚ท, ์•ผ์™ธ, ๊ฑฐ๋ฆฌ, ์†Œ๋…€

sample-500 Test prompt: 1์†Œ๋…€๊ฐ€ ๊ธฐํƒ€ ์—ฐ์ฃผ๋ฅผ ํ•˜๊ณ  ์žˆ์Šต๋‹ˆ๋‹ค, ๋ฐ”๋‹ค, ๋ˆˆ์„ ๊ฐ์Œ, ๊ธด ๋จธ๋ฆฌ, ์นด๋ฆฌ์Šค๋งˆ

sample-500 Test prompt: 1์†Œ๋…„, ๊ธฐํƒ€ ์—ฐ์ฃผ, ์ž˜์ƒ๊น€, ์•‰์•„์žˆ๋Š”, ๋นจ๊ฐ„์ƒ‰ ๊ธฐํƒ€, ํ•ด๋ณ€

Usage

Clone the github repo

git clone https://github.com/showlab/Tune-A-Video.git

Run inference code

from tuneavideo.pipelines.pipeline_tuneavideo import TuneAVideoPipeline
from tuneavideo.models.unet import UNet3DConditionModel
from tuneavideo.util import save_videos_grid
import torch

pretrained_model_path = "kyujinpy/KO-anything-v4-5"
unet_model_path = "kyujinpy/Tune-A-VideKO-anything"
unet = UNet3DConditionModel.from_pretrained(unet_model_path, subfolder='unet', torch_dtype=torch.float16).to('cuda')
pipe = TuneAVideoPipeline.from_pretrained(pretrained_model_path, unet=unet, torch_dtype=torch.float16).to("cuda")
pipe.enable_xformers_memory_efficient_attention()

prompt = "1์†Œ๋…€๋Š” ๊ธฐํƒ€๋ฅผ ์—ฐ์ฃผํ•˜๊ณ  ์žˆ๋‹ค, ํฐ ๋จธ๋ฆฌ, ์ค‘๊ฐ„ ๋จธ๋ฆฌ, ๊ณ ์–‘์ด ๊ท€, ๊ท€์—ฌ์šด, ์Šค์นดํ”„, ์žฌํ‚ท, ์•ผ์™ธ, ๊ฑฐ๋ฆฌ, ์†Œ๋…€"
video = pipe(prompt, video_length=14, height=512, width=512, num_inference_steps=50, guidance_scale=7.5).videos

save_videos_grid(video, f"./{prompt}.gif")

Related Papers:

  • Tune-A-Video: One-Shot Tuning of Image Diffusion Models for Text-to-Video Generation
  • Stable Diffusion: High-Resolution Image Synthesis with Latent Diffusion Models