Gemini Powered Vision App ( Image Analyzer)

Overview

This project is a Streamlit-based application that allows users to upload images and ask questions about them using the Gemini AI model. The application leverages generative AI to provide insightful responses based on the uploaded images and user queries.

Features

• Image Upload: Users can upload images in various formats (PNG, JPG, WEBP, JPEG).
• Interactive Q&A: Engage with the AI by asking questions about the uploaded image.
• User-Friendly Design: A simple and intuitive interface built with Streamlit.
• Environment Variable Management: Securely manage API keys using dotenv.

Technologies Used

• Python: The primary programming language for development, focused on AI models.
• Streamlit: For building the user interface.
• Google Gemini API: For generating responses based on images and queries.
• dotenv: For managing environment variables.
• Pillow: For image processing.

Getting Started

Dependencies

This code uses the following libraries:

• streamlit: for building the user interface.
• google.generativeai: for generating responses using the Gemini model.
• python-dotenv: for managing environment variables.
• Pillow: for handling image uploads and processing.

Contributing

We welcome contributions to enhance the Gemini PicAnalyzer! Here’s how you can get involved:

1. Fork the Repository: Click on the "Fork" button at the top right of this page.
2. Clone Your Fork: Clone your forked repository to your local machine.

   git clone https://github.com/AlexKalll/Gemini-Vision-App.git

Usage

Follow these steps to set up and run the project:

1. Clone the repository:

   git clone https://github.com/AlexKalll/Gemini-Vision-App.git

2. Create a virtual environment:

   python3 -m venv my_env
   source my_env/bin/activate 
   # For Windows
   .\my_env\Scripts\activate 

3. Install dependencies:

   pip install -r requirements.txt

4. Run the Streamlit server:

   streamlit run main.py

5. Access the application in your browser at http://localhost:8501.

6. Start analyzing images with the assistant!

7. Acces the Source Code

Repository Structure

repository/
├── main.py                       # The code and UI integrated together live here
├── requirements.txt              # The Python packages needed to run locally
├── Demo.mp4                      # Demo of the application
├── README.md                     # Documentation for the project
├── .gitignore                    # Files to ignore in the repository

How it Works

1. The user uploads an image using the file uploader.
2. The user enters a question related to the uploaded image.
3. The input query and image are sent to the Gemini model for processing.
4. The Gemini model generates a response based on the image and the user's question.
5. The application displays the response to the user.

Code Explanation

Here’s a brief overview of the main components of the code in main.py:

import os 
import streamlit as st
import google.generativeai as genai
from PIL import Image
from dotenv import load_dotenv

load_dotenv()
my_api_key = os.getenv("GOOGLE_API_KEY")
genai.configure(api_key=my_api_key)
model = genai.GenerativeModel('gemini-1.5-flash')

def get_gemini_response(user_quest, image):
    response = model.generate_content([user_quest, image])
    return response.text

st.set_page_config(page_title='GeminiVision App', page_icon='🤖')
st.title("Gemini PicAnalyzer Application ")
file = st.file_uploader('Choose an image...', type=['png', 'jpg', 'webp'])

if file is not None:
    img = Image.open(file)
    st.image(img, width=200)

st.subheader("Ask the model about the image")  
with st.form(key='input', clear_on_submit=True):
    user_quest = st.text_input("Enter Your question about an image you uploaded.", placeholder='Type here...')
    btn = st.form_submit_button()

if btn and user_quest:
    response = get_gemini_response(user_quest, img)
    st.subheader("Answer: ")
    st.write(response)
else:
    st.markdown('<div style="height: 150px;"></div>', unsafe_allow_html=True) 
    st.markdown('---')
    st.success("Made By Kaletsidik ©2024")

Let's Get Introduced

Feel free to reach out for any inquiries or feedback:

• LinkedIn
• Instagram
• X (formerly Twitter)
• GitHub
• LeetCode

Kaletsidik Ayalew
alexkalalw@gmail.com