app.py

import streamlit as st
from PIL import Image
from deepface import DeepFace
import cv2
import numpy as np


# Set background color to black
st.markdown(
    """
    <style>
    body {
        background-color: #000000;
        color: white;
    }
    </style>
    """,
    unsafe_allow_html=True
)

# Center-align the heading
st.markdown(
    """
    <h1 style='text-align: center;'>DeepFace - Age, Gender, Race, Expression Detection Program</h1>
    """,
    unsafe_allow_html=True
)

# Add the description
st.markdown(
    """
    <p style='text-align: center;'>Application Created by:<br>
     Shahzad Ahmed (2K20/IT/114)<br>
     Aziz Dil Khan (2K20/IT/31)<br>
     Muhammad Ahsan Qureshi (2K20/IT/68)</p>
    """,
    unsafe_allow_html=True
)

def drawFace(image, results):
    for result in results:
        x = result['region']['x']
        y = result['region']['y']
        h = result['region']['h']
        w = result['region']['w']
        age = result['age']
        gender = 'M' if result['gender']['Man'] > result['gender']['Woman'] else 'F'
        dominant_emotion = result['dominant_emotion']
        dominant_race = result['dominant_race']
        description = f'{age}{gender} - {dominant_emotion}'

        cv2.rectangle(image, (x, y), (x + w, y + h), (0, 255, 0), 2)
        cv2.putText(image, description, (x, y - 10), cv2.FONT_HERSHEY_PLAIN, 2, (0, 255, 0), 3)
        cv2.putText(image, dominant_race, (x, y + h + 30), cv2.FONT_HERSHEY_PLAIN, 2, (0, 255, 0), 3)

    return image
st.title("DeepFace Analysis")

# Add file upload option
uploaded_file = st.file_uploader("Upload an image", type=["jpg", "jpeg", "png"])

if uploaded_file is not None:
    # Read the uploaded image
    image = Image.open(uploaded_file)
    img_path = "temp_image.jpg"  # Save the image temporarily

    # Save the image temporarily for OpenCV processing
    image.save(img_path)

    # Analyze the image using DeepFace
    results = DeepFace.analyze(img_path=img_path, actions=['age', 'gender', 'race', 'emotion'])


    # Read the image using OpenCV
    cv_image = cv2.imread(img_path)
    # Convert BGR to RGB color channel ordering
    cv_image_rgb = cv2.cvtColor(cv_image, cv2.COLOR_BGR2RGB)

    # Adjust color balance
    cv_image_rgb_adjusted = cv2.cvtColor(cv_image_rgb, cv2.COLOR_RGB2LAB)
    avg_a = np.average(cv_image_rgb_adjusted[:, :, 1])
    avg_b = np.average(cv_image_rgb_adjusted[:, :, 2])
    cv_image_rgb_adjusted[:, :, 1] = cv_image_rgb_adjusted[:, :, 1] - ((np.average(cv_image_rgb_adjusted[:, :, 1]) - 128) * (cv_image_rgb_adjusted[:, :, 0] / 255.0) * 1.1)
    cv_image_rgb_adjusted[:, :, 2] = cv_image_rgb_adjusted[:, :, 2] - ((np.average(cv_image_rgb_adjusted[:, :, 2]) - 128) * (cv_image_rgb_adjusted[:, :, 0] / 255.0) * 1.1)
    cv_image_rgb_adjusted = cv2.cvtColor(cv_image_rgb_adjusted, cv2.COLOR_LAB2RGB)

    annotated_image = drawFace(cv_image_rgb_adjusted, results)


    # Display image and text side by side
    col1, col2 = st.columns(2)
    col1.image(annotated_image, caption='Annotated Image', use_column_width=True)
    col2.write("Results:")
    col2.write(results)