Created and implemented Overlay face function

[tsu-ki]

Closes: #2527
#3243
I've also made changes to delete_issue function to reduce time complexity and better optimisation

Screen.Recording.2025-01-19.at.6.10.46.PM.mov

[DonnieBLT]

I'm concerned about the memory usage of numpy, is there a way to know how much more memory this will consume?

[tsu-ki]

@DonnieBLT sir, I'm unsure of why pre-commit test is failing continuously even though I'm running pre-commit everytime before committing the changes. Please suggest what should I do to resolve this, thank you!

[tsu-ki]

Collaborator

I've added tracemalloc to track and display the memory consumption during execution of overlay_function. I feel the memory consumption won't increase significantly but if needed, I can make further changes to decrease it.

[DonnieBLT]

Can you please add an automated test to this so that it starts with a face, then upload it to an issue and then check the issue and make sure that it does not detect a face

[DonnieBLT]

@DonnieBLT sir, I'm unsure of why pre-commit test is failing continuously even though I'm running pre-commit everytime before committing the changes. Please suggest what should I do to resolve this, thank you!

It’s precommit formatting

[tsu-ki]

Can you please add an automated test to this so that it starts with a face, then upload it to an issue and then check the issue and make sure that it does not detect a face

sure sir, I'll do that. thank you!

[tsu-ki]

Can you please add an automated test to this so that it starts with a face, then upload it to an issue and then check the issue and make sure that it does not detect a face

@DonnieBLT sir, can you please explain what was required to be done here?

[DonnieBLT]

Can you please add an automated test to this so that it starts with a face, then upload it to an issue and then check the issue and make sure that it does not detect a face

@DonnieBLT sir, can you please explain what was required to be done here?

Create a test

[tsu-ki]

@DonnieBLT sir, please review the changes. thank you!

Screen.Recording.2025-02-05.at.5.46.25.PM.mov

[DonnieBLT]

Please remove unused, dependencies, and separate the functionality into a new PR if it’s still there

[tsu-ki]

@DonnieBLT
before:

after downloading dependencies: (opencv-python, numpy):

[DonnieBLT]

lets do this without 3rd party dependencies - something like this

import os
import urllib.request
from xml.etree import ElementTree
from PIL import Image, ImageDraw

# Load the Haar cascade file (pre-trained model for face detection)
HAAR_CASCADE_URL = "https://raw.githubusercontent.com/opencv/opencv/master/data/haarcascades/haarcascade_frontalface_default.xml"
HAAR_CASCADE_PATH = "haarcascade_frontalface_default.xml"

# Download Haar cascade if not exists
if not os.path.exists(HAAR_CASCADE_PATH):
    urllib.request.urlretrieve(HAAR_CASCADE_URL, HAAR_CASCADE_PATH)

# Parse Haar cascade XML file
def parse_haar_cascade(xml_path):
    """Parses Haar cascade XML and extracts feature rectangles."""
    tree = ElementTree.parse(xml_path)
    root = tree.getroot()
    features = []
    for stage in root.findall(".//stages/_/trees/_/feature/rects/_"):
        feature_rects = []
        for rect in stage.text.strip().split("\n"):
            rect = rect.split()
            x, y, w, h = map(int, rect[:4])
            feature_rects.append((x, y, w, h))
        features.append(feature_rects)
    return features

# Face detection function using Haar features
def detect_faces(image):
    """Detect faces in an image using Haar features."""
    width, height = image.size
    features = parse_haar_cascade(HAAR_CASCADE_PATH)
    detected_faces = []
    
    for x in range(0, width - 24, 2):  # Scanning the image
        for y in range(0, height - 24, 2):
            for feature in features:
                if all(x + w < width and y + h < height for (_, _, w, h) in feature):
                    detected_faces.append((x, y, 24, 24))  # Assume minimum size of 24x24 pixels
    return detected_faces

# Function to blur detected faces
def blur_faces(image, faces):
    """Apply a blur effect on detected faces."""
    draw = ImageDraw.Draw(image)
    for x, y, w, h in faces:
        for i in range(5):  # Create blur effect using rectangles
            draw.rectangle([x + i, y + i, x + w - i, y + h - i], fill=(200, 200, 200, 128))
    return image

# Load the image
image_path = "test_image.jpg"  # Replace with your image path
image = Image.open(image_path).convert("RGB")

# Detect faces
faces = detect_faces(image)

# Apply blur
processed_image = blur_faces(image, faces)

# Save or show the processed image
processed_image.show()  # Display the image
processed_image.save("blurred_faces.jpg")  # Save the processed image