File size: 6,686 Bytes
9c99365 |
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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 |
import streamlit as st
import requests
import base64
from io import BytesIO
from PIL import Image
import time
# Streamlit app title and description
st.title("SDXL Image Generation Wrapper")
st.markdown("""
This Streamlit app wraps the Google SDXL Gradio app to generate images from text prompts.
Enter a prompt, select a style, and adjust settings to create high-quality images.
""")
# Backend URL of the Gradio app (replace with actual endpoint if different)
BACKEND_URL = "https://google-sdxl.hf.space/api/predict/"
# Style options (mirroring the Gradio app's style_list)
STYLE_OPTIONS = [
"(No style)", "Cinematic", "Photographic", "Anime", "Manga",
"Digital Art", "Pixel art", "Fantasy art", "Neonpunk", "3D Model"
]
# Function to apply style (replicating the Gradio app's apply_style logic)
def apply_style(style_name, prompt, negative):
style_dict = {
"(No style)": ("{prompt}", ""),
"Cinematic": (
"cinematic still {prompt} . emotional, harmonious, vignette, highly detailed, high budget, bokeh, cinemascope, moody, epic, gorgeous, film grain, grainy",
"anime, cartoon, graphic, text, painting, crayon, graphite, abstract, glitch, deformed, mutated, ugly, disfigured"
),
"Photographic": (
"cinematic photo {prompt} . 35mm photograph, film, bokeh, professional, 4k, highly detailed",
"drawing, painting, crayon, sketch, graphite, impressionist, noisy, blurry, soft, deformed, ugly"
),
"Anime": (
"anime artwork {prompt} . anime style, key visual, vibrant, studio anime, highly detailed",
"photo, deformed, black and white, realism, disfigured, low contrast"
),
"Manga": (
"manga style {prompt} . vibrant, high-energy, detailed, iconic, Japanese comic style",
"ugly, deformed, noisy, blurry, low contrast, realism, photorealistic, Western comic style"
),
"Digital Art": (
"concept art {prompt} . digital artwork, illustrative, painterly, matte painting, highly detailed",
"photo, photorealistic, realism, ugly"
),
"Pixel art": (
"pixel-art {prompt} . low-res, blocky, pixel art style, 8-bit graphics",
"sloppy, messy, blurry, noisy, highly detailed, ultra textured, photo, realistic"
),
"Fantasy art": (
"ethereal fantasy concept art of {prompt} . magnificent, celestial, ethereal, painterly, epic, majestic, magical, fantasy art, cover art, dreamy",
"photographic, realistic, realism, 35mm film, dslr, cropped, frame, text, deformed, glitch, noise, noisy, off-center, deformed, cross-eyed, closed eyes, bad anatomy, ugly, disfigured, sloppy, duplicate, mutated, black and white"
),
"Neonpunk": (
"neonpunk style {prompt} . cyberpunk, vaporwave, neon, vibes, vibrant, stunningly beautiful, crisp, detailed, sleek, ultramodern, magenta highlights, dark purple shadows, high contrast, cinematic, ultra detailed, intricate, professional",
"painting, drawing, illustration, glitch, deformed, mutated, cross-eyed, ugly, disfigured"
),
"3D Model": (
"professional 3d model {prompt} . octane render, highly detailed, volumetric, dramatic lighting",
"ugly, deformed, noisy, low poly, blurry, painting"
)
}
positive, neg = style_dict.get(style_name, style_dict["(No style)"])
return positive.replace("{prompt}", prompt), neg + (negative or "")
# Function to call the Gradio backend
def generate_images(prompt, negative_prompt, guidance_scale, style_name):
try:
# Apply style to prompt and negative prompt
styled_prompt, styled_negative = apply_style(style_name, prompt, negative_prompt)
# Prepare payload (mimicking Gradio's infer function)
payload = {
"data": [
styled_prompt,
styled_negative,
guidance_scale,
style_name
]
}
# Send request to Gradio backend
start_time = time.time()
response = requests.post(BACKEND_URL, json=payload, timeout=60)
elapsed_time = time.time() - start_time
# Check response
if response.status_code != 200:
st.error(f"Error: Received status code {response.status_code}")
return None
# Parse response
json_data = response.json()
if "data" not in json_data or not json_data["data"]:
st.error("Error: No images returned from the backend.")
return None
# Extract images (assuming base64 strings)
images = []
for img_data in json_data["data"][0]: # Adjust based on actual response structure
if isinstance(img_data, str) and img_data.startswith("data:image"):
img_base64 = img_data.split(",")[1]
img_bytes = base64.b64decode(img_base64)
img = Image.open(BytesIO(img_bytes))
images.append(img)
else:
st.warning("Unexpected image data format.")
st.success(f"Images generated in {elapsed_time:.2f} seconds!")
return images
except requests.exceptions.RequestException as e:
st.error(f"Network error: {str(e)}")
return None
except ValueError as e:
st.error(f"Error decoding response: {str(e)}")
return None
except Exception as e:
st.error(f"Unexpected error: {str(e)}")
return None
# Streamlit UI components
with st.form(key="input_form"):
prompt = st.text_input("Enter your prompt", placeholder="A serious capybara at work, wearing a suit")
negative_prompt = st.text_input("Negative prompt (optional)", placeholder="low_quality")
guidance_scale = st.slider("Guidance Scale", min_value=0.0, max_value=50.0, value=7.5, step=0.1)
style_name = st.selectbox("Image Style", options=STYLE_OPTIONS, index=0)
submit_button = st.form_submit_button("Generate Images")
# Handle form submission
if submit_button:
if not prompt:
st.error("Please enter a prompt.")
else:
with st.spinner("Generating images..."):
images = generate_images(prompt, negative_prompt, guidance_scale, style_name)
if images:
# Display images in a gallery
st.subheader("Generated Images")
cols = st.columns(min(len(images), 3)) # Adjust columns based on number of images
for idx, img in enumerate(images):
with cols[idx % 3]:
st.image(img, use_column_width=True) |