app.py

#!/usr/bin/env python

import os
import pathlib
import tempfile

import gradio as gr
import torch
from huggingface_hub import snapshot_download
from modelscope.outputs import OutputKeys
from modelscope.pipelines import pipeline

DESCRIPTION = "# ModelScope-Image2Video"
if not torch.cuda.is_available():
    DESCRIPTION += "\n<p>Running on CPU 🥶 This demo does not work on CPU.</p>"

if torch.cuda.is_available():
    model_cache_dir = os.getenv("MODEL_CACHE_DIR", "./models")
    model_dir = pathlib.Path(model_cache_dir) / "MS-Image2Video"
    snapshot_download(repo_id="damo-vilab/MS-Image2Video", repo_type="model", local_dir=model_dir)
    pipe = pipeline(task="image-to-video", model=model_dir.as_posix(), model_revision="v1.1.0", device="cuda:0")
else:
    pipe = None


def image_to_video(image_path: str) -> str:
    output_file = tempfile.NamedTemporaryFile(suffix=".mp4", delete=False)
    pipe(image_path, output_video=output_file.name)[OutputKeys.OUTPUT_VIDEO]
    return output_file.name


with gr.Blocks(css="style.css") as demo:
    gr.Markdown(DESCRIPTION)
    gr.DuplicateButton(
        value="Duplicate Space for private use",
        elem_id="duplicate-button",
        visible=os.getenv("SHOW_DUPLICATE_BUTTON") == "1",
    )
    with gr.Group():
        input_image = gr.Image(label="Input image", type="filepath")
        run_button = gr.Button()
        output_video = gr.Video(label="Output video")

    run_button.click(
        fn=image_to_video,
        inputs=input_image,
        outputs=output_video,
        api_name="run",
    )

if __name__ == "__main__":
    demo.queue(max_size=10).launch()