Spaces:
Running
Running
File size: 1,959 Bytes
1e350c6 0f09deb 1e350c6 0f09deb 1e350c6 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 |
import os
import streamlit as st
from diffusers import CogVideoXImageToVideoPipeline
from diffusers.utils import export_to_video, load_image
import torch
# Streamlit interface for uploading an image and inputting a prompt
st.title("Image to Video with Hugging Face")
st.write("Upload an image and provide a prompt to generate a video.")
# File uploader for the input image
uploaded_file = st.file_uploader("Upload an image (JPG or PNG):", type=["jpg", "jpeg", "png"])
prompt = st.text_input("Enter your prompt:", "A little girl is riding a bicycle at high speed. Focused, detailed, realistic.")
if uploaded_file and prompt:
try:
# Save the uploaded file to a temporary location
with open("uploaded_image.jpg", "wb") as f:
f.write(uploaded_file.read())
# Load the image
image = load_image("uploaded_image.jpg")
# Initialize the CogVideoX pipeline
st.write("Initializing the pipeline...")
pipe = CogVideoXImageToVideoPipeline.from_pretrained(
"THUDM/CogVideoX1.5-5B-I2V",
torch_dtype=torch.bfloat16
)
pipe.enable_sequential_cpu_offload()
pipe.vae.enable_tiling()
pipe.vae.enable_slicing()
# Generate the video
st.write("Generating video... this may take a while.")
video_frames = pipe(
prompt=prompt,
image=image,
num_videos_per_prompt=1,
num_inference_steps=50,
num_frames=81,
guidance_scale=6,
generator=torch.Generator(device="cuda").manual_seed(42),
).frames[0]
# Export the video
video_path = "output.mp4"
export_to_video(video_frames, video_path, fps=8)
# Display the video in Streamlit
st.video(video_path)
except Exception as e:
st.error(f"An error occurred: {e}")
else:
st.write("Please upload an image and provide a prompt to get started.")
|