app.py

import gradio as gr
from transformers import pipeline
from PIL import Image
import requests
from transformers import BlipProcessor, BlipForConditionalGeneration

# Initialize the pipeline
pipe = pipeline("image-to-text", model="Salesforce/blip-image-captioning-large")

# Initialize processor and model
processor = BlipProcessor.from_pretrained("Salesforce/blip-image-captioning-large")
model = BlipForConditionalGeneration.from_pretrained("Salesforce/blip-image-captioning-large")

def image_caption(image, text_prompt=None):
    # Conditional image captioning if text prompt is provided
    if text_prompt:
        inputs = processor(image, text_prompt, return_tensors="pt")
        out = model.generate(**inputs)
        caption = processor.decode(out[0], skip_special_tokens=True)
    else:
        # Unconditional image captioning
        inputs = processor(image, return_tensors="pt")
        out = model.generate(**inputs)
        caption = processor.decode(out[0], skip_special_tokens=True)
    return caption

# Define the Gradio interface
image_input = gr.Image(type="pil", label="Upload an Image")
text_input = gr.Textbox(lines=1, placeholder="Optional: Enter text prompt", label="Text Prompt")
output = gr.Textbox(label="Generated Caption")

gr.Interface(
    fn=image_caption,
    inputs=[image_input, text_input],
    outputs=output,
    title="Image Captioning with BLIP",
    description="Upload an image and get a generated caption. Optionally, provide a text prompt for conditional captioning."
).launch()