Spaces:
Running
Running
| import streamlit as st | |
| import requests | |
| import base64 | |
| import os | |
| # Function to convert image to base64 | |
| def convert_image_to_base64(image): | |
| image_bytes = image.read() | |
| encoded_image = base64.b64encode(image_bytes).decode("utf-8") | |
| return encoded_image | |
| # Function to generate a caption using Nebius API | |
| def generate_caption(encoded_image): | |
| API_URL = "https://api.studio.nebius.ai/v1/chat/completions" | |
| API_KEY = os.environ.get("NEBIUS_API_KEY") | |
| headers = { | |
| "Authorization": f"Bearer {API_KEY}", | |
| "Content-Type": "application/json" | |
| } | |
| payload = { | |
| "model": "Qwen/Qwen2-VL-7B-Instruct", | |
| "messages": [ | |
| { | |
| "role": "system", | |
| "content": """You are an image to prompt converter. Your work is to observe each and every detail of the image and craft a detailed prompt under 75 words in this format: [image content/subject, description of action, state, and mood], [art form, style], [artist/photographer reference if needed], [additional settings such as camera and lens settings, lighting, colors, effects, texture, background, rendering].""" | |
| }, | |
| { | |
| "role": "user", | |
| "content": "write a detailed caption for this image" | |
| } | |
| ], | |
| "image": { | |
| "type": "image_url", | |
| "image_url": { | |
| "url": f"data:image/png;base64,{encoded_image}" | |
| } | |
| }, | |
| "temperature": 0 | |
| } | |
| response = requests.post(API_URL, headers=headers, json=payload) | |
| if response.status_code == 200: | |
| result = response.json() | |
| caption = result.get("choices", [{}])[0].get("message", {}).get("content", "No caption generated.") | |
| return caption | |
| else: | |
| st.error(f"API Error {response.status_code}: {response.text}") | |
| return None | |
| # Streamlit app | |
| def main(): | |
| st.set_page_config(page_title="Image to Caption Converter", layout="centered", initial_sidebar_state="collapsed") | |
| st.title("🖼️ Image to Caption Converter") | |
| st.markdown("Upload an image and let the AI generate a detailed caption for it.") | |
| uploaded_file = st.file_uploader("Upload an image", type=["jpg", "jpeg", "png"]) | |
| if uploaded_file: | |
| # Display the uploaded image | |
| st.image(uploaded_file, caption="Uploaded Image", use_container_width=True) | |
| # Convert image to base64 and get caption | |
| if st.button("Generate Caption"): | |
| with st.spinner("Generating caption..."): | |
| encoded_image = convert_image_to_base64(uploaded_file) | |
| caption = generate_caption(encoded_image) | |
| if caption: | |
| st.subheader("Generated Caption:") | |
| st.text_area("", caption, height=100, key="caption_area") | |
| # Copy button | |
| st.button("Copy to Clipboard", on_click=lambda: st.session_state.update({"caption": caption})) | |
| if __name__ == "__main__": | |
| main() | |