app.py

import streamlit as st
from transformers import pipeline
from PIL import Image




st.header("Character Captions")
st.write("Have a character caption any image you upload!")
character = st.selectbox("Choose a character", ["rapper", "monkey", "shrek", "unintelligible"])

uploaded_img = st.file_uploader("Upload an image")

if uploaded_img is not None:

    image = Image.open(uploaded_img)
    st.image(image)

    image_captioner = pipeline("image-to-text", model="Salesforce/blip-image-captioning-large")

    response = image_captioner(image)
    caption = response[0]['generated_text']
    st.write("Caption:", caption)

    character_prompts = {
        "rapper": f"Describe this scene like you're a rapper singing: {caption}.",
        "monkey": f"Describe this scene like you're a monkey going bananas: {caption}.",
        "shrek": f"Describe this scene like you're Shrek: {caption}.",
        "unintelligible": f"Describe this scene in a way that makes no sense: {caption}."
    }

    prompt = character_prompts[character]
    st.write(prompt)




# pipe = pipeline("text-generation", model="meta-llama/Meta-Llama-Guard-2-8B")

# # Pass the caption to another model

# personality = "rapper"
# prompt = personality_prompts[personality]

# styled_caption = text_generator(prompt, max_length=100, num_return_sequences=1)

# print(styled_caption[0]["generated_text"])