# app.py import os import io import base64 from PIL import Image as PImage from transformers import pipeline import gradio as gr from dotenv import load_dotenv # pip install python-dotenv # ------------------------------- # 1️⃣ Load environment variables from .env # ------------------------------- load_dotenv() hf_api_key = os.getenv("HF_API_KEY") if not hf_api_key: raise ValueError("HF_API_KEY not found in .env file") os.environ["HF_API_KEY"] = hf_api_key # ------------------------------- # 2️⃣ Initialize BLIP pipeline # ------------------------------- get_completion = pipeline("image-to-text", model="Salesforce/blip-image-captioning-base") # ------------------------------- # 3️⃣ Helper function: PIL -> base64 # ------------------------------- def image_to_base64_str(pil_image): byte_arr = io.BytesIO() pil_image.save(byte_arr, format='PNG') byte_arr = byte_arr.getvalue() return base64.b64encode(byte_arr).decode('utf-8') # ------------------------------- # 4️⃣ Generate caption from base64 image # ------------------------------- def generate_caption(base64_image): image_data = base64.b64decode(base64_image) image = PImage.open(io.BytesIO(image_data)) caption_result = get_completion(image) if isinstance(caption_result, list) and len(caption_result) > 0: return caption_result[0].get('generated_text', "No caption generated") else: return "No caption generated" # ------------------------------- # 5️⃣ Captioner function for Gradio # ------------------------------- def captioner(image): base64_image = image_to_base64_str(image) caption_text = generate_caption(base64_image) print("Caption:", caption_text) return caption_text # ------------------------------- # 6️⃣ Example images (must be in project folder) # ------------------------------- examples = ["christmas_dog.jpeg", "bird_flight.jpeg", "cow.jpeg", "dog_animal_greyhound.jpeg"] # ------------------------------- # 7️⃣ Gradio Interface # ------------------------------- demo = gr.Interface( fn=captioner, inputs=[gr.Image(label="Upload Image", type="pil")], outputs=[gr.Textbox(label="Caption")], title="Image Captioning with BLIP", description="Caption any image using the BLIP model", allow_flagging="never", examples=examples ) # Launch the app demo.launch(share=True, server_port=int(os.environ.get('PORT', 7860)))