app.py

import streamlit as st
from PIL import Image
import torch

from transformers import Blip2Processor, Blip2ForConditionalGeneration

processor = Blip2Processor.from_pretrained("Salesforce/blip2-opt-2.7b")
model = Blip2ForConditionalGeneration.from_pretrained(
    "Salesforce/blip2-opt-2.7b", device_map={"": 0}, torch_dtype=torch.float16
)


enable = st.checkbox("Enable camera")
picture = st.camera_input("Take a picture", disabled=not enable)

if picture:
	image = Image.open(picture)
	prompt = "Question: At what location is this person most likely attending this online meeting? Answer:"
	inputs = processor(images=image, text=prompt, return_tensors="pt").to(device="cuda", dtype=torch.float16)

	generated_ids = model.generate(**inputs)
	generated_text = processor.batch_decode(generated_ids, skip_special_tokens=True)[0].strip()
	st.write(generated_text)