Spaces:
Runtime error
Runtime error
File size: 4,610 Bytes
cb5a657 18ed7af cb5a657 1c5196e cb5a657 2927d16 cb5a657 ebdbb36 cb5a657 18ed7af cb5a657 |
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 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 |
import gradio as gr
#import gradio.helpers
import torch
import os
import base64
from glob import glob
from pathlib import Path
from typing import Optional
from diffusers import StableVideoDiffusionPipeline
from diffusers.utils import load_image, export_to_video
from PIL import Image
import uuid
import random
from huggingface_hub import login, hf_hub_download
import spaces
pipe = StableVideoDiffusionPipeline.from_pretrained(
# "stabilityai/stable-video-diffusion-img2vid-xt-1-1",
"vdo/stable-video-diffusion-img2vid-xt-1-1",
torch_dtype=torch.float16,
variant="fp16"
)
pipe.save_pretrained("model", variant="fp16")
pipe.to("cuda")
pipe.unet = torch.compile(pipe.unet, mode="reduce-overhead", fullgraph=True)
#pipe.vae = torch.compile(pipe.vae, mode="reduce-overhead", fullgraph=True)
max_64_bit_int = 2**63 - 1
@spaces.GPU(enable_queue=True)
def generate_video(
image: Image,
seed: int,
motion_bucket_id: int = 127,
fps_id: int = 6,
version: str = "svd_xt",
cond_aug: float = 0.02,
decoding_t: int = 3, # Number of frames decoded at a time! This eats most VRAM. Reduce if necessary.
device: str = "cuda",
output_folder: str = "outputs",
):
# note julian: normally we should resize input images, but normally they are already in 1024x576, so..
# also, I would like to experiment with vertical videos, and 1024x512 videos
image = resize_image(image)
if image.mode == "RGBA":
image = image.convert("RGB")
generator = torch.manual_seed(seed)
os.makedirs(output_folder, exist_ok=True)
base_count = len(glob(os.path.join(output_folder, "*.mp4")))
video_path = os.path.join(output_folder, f"{base_count:06d}.mp4")
frames = pipe(image, decode_chunk_size=decoding_t, generator=generator, motion_bucket_id=motion_bucket_id, noise_aug_strength=0.1, num_frames=25).frames[0]
export_to_video(frames, video_path, fps=fps_id)
torch.manual_seed(seed)
# Read the content of the video file and encode it to base64
with open(video_path, "rb") as video_file:
video_base64 = base64.b64encode(video_file.read()).decode('utf-8')
# Prepend the appropriate data URI header with MIME type
video_data_uri = 'data:video/mp4;base64,' + video_base64
# clean-up (otherwise there is a risk of "ghosting", eg. someone seeing the previous generated video",
# of one of the steps go wrong)
os.remove(video_path)
return video_data_uri
def resize_image(image, output_size=(1024, 576)):
# Calculate aspect ratios
target_aspect = output_size[0] / output_size[1] # Aspect ratio of the desired size
image_aspect = image.width / image.height # Aspect ratio of the original image
# Resize then crop if the original image is larger
if image_aspect > target_aspect:
# Resize the image to match the target height, maintaining aspect ratio
new_height = output_size[1]
new_width = int(new_height * image_aspect)
resized_image = image.resize((new_width, new_height), Image.LANCZOS)
# Calculate coordinates for cropping
left = (new_width - output_size[0]) / 2
top = 0
right = (new_width + output_size[0]) / 2
bottom = output_size[1]
else:
# Resize the image to match the target width, maintaining aspect ratio
new_width = output_size[0]
new_height = int(new_width / image_aspect)
resized_image = image.resize((new_width, new_height), Image.LANCZOS)
# Calculate coordinates for cropping
left = 0
top = (new_height - output_size[1]) / 2
right = output_size[0]
bottom = (new_height + output_size[1]) / 2
# Crop the image
cropped_image = resized_image.crop((left, top, right, bottom))
return cropped_image
with gr.Blocks() as demo:
image = gr.Image(label="Upload your image", type="pil")
generate_btn = gr.Button("Generate")
base64_out = gr.Textbox(label="Base64 Video")
seed = gr.Slider(label="Seed", value=42, randomize=False, minimum=0, maximum=max_64_bit_int, step=1)
motion_bucket_id = gr.Slider(label="Motion bucket id", info="Controls how much motion to add/remove from the image", value=127, minimum=1, maximum=255)
fps_id = gr.Slider(label="Frames per second", info="The length of your video in seconds will be 25/fps", value=6, minimum=5, maximum=30)
generate_btn.click(
fn=generate_video,
inputs=[image, seed, motion_bucket_id, fps_id],
outputs=base64_out,
api_name="run"
)
demo.queue(max_size=20).launch() |