Computer Vision

Computer Vision

Overview

Computer vision enables machines to understand visual information from images and videos, powering applications like autonomous driving, medical imaging, and surveillance.

When to Use

• Image classification and object recognition tasks
• Object detection and localization in images
• Semantic or instance segmentation projects
• Pose estimation and human activity recognition
• Face recognition and biometric systems
• Medical imaging analysis and diagnostics

Computer Vision Tasks

• Image Classification: Categorizing images into classes
• Object Detection: Locating and classifying objects in images
• Semantic Segmentation: Pixel-level classification
• Instance Segmentation: Detecting individual object instances
• Pose Estimation: Identifying human body joints
• Face Recognition: Identifying individuals in images

Popular Architectures

• Classification: ResNet, VGG, EfficientNet, Vision Transformer
• Detection: YOLO, Faster R-CNN, SSD, RetinaNet
• Segmentation: U-Net, DeepLab, Mask R-CNN
• Pose: OpenPose, PoseNet, HRNet

Python Implementation

import numpy as np
import pandas as pd
import matplotlib.pyplot as plt
import matplotlib.patches as patches
from PIL import Image, ImageDraw
import torch
import torch.nn as nn
from torch.utils.data import DataLoader, TensorDataset
from torchvision import transforms, models, datasets
import tensorflow as tf
from tensorflow import keras
from tensorflow.keras import layers
import cv2
from sklearn.metrics import accuracy_score, confusion_matrix
import seaborn as sns
import warnings
warnings.filterwarnings('ignore')

print("=== 1. Image Classification CNN ===")

# Define image classification model
class ImageClassifierCNN(nn.Module):
    def __init__(self, num_classes=10):
        super().__init__()
        self.features = nn.Sequential(
            nn.Conv2d(3, 32, kernel_size=3, padding=1),
            nn.ReLU(inplace=True),
            nn.BatchNorm2d(32),
            nn.MaxPool2d(2, 2),

            nn.Conv2d(32, 64, kernel_size=3, padding=1),
            nn.ReLU(inplace=True),
            nn.BatchNorm2d(64),
            nn.MaxPool2d(2, 2),

            nn.Conv2d(64, 128, kernel_size=3, padding=1),
            nn.ReLU(inplace=True),
            nn.BatchNorm2d(128),
            nn.MaxPool2d(2, 2),
        )

        self.classifier = nn.Sequential(
            nn.Linear(128 * 4 * 4, 256),
            nn.ReLU(inplace=True),
            nn.Dropout(0.5),
            nn.Linear(256, num_classes)
        )

    def forward(self, x):
        x = self.features(x)
        x = x.view(x.size(0), -1)
        x = self.classifier(x)
        return x

model = ImageClassifierCNN(num_classes=10)
print(f"Model parameters: {sum(p.numel() for p in model.parameters()):,}")

# 2. Object Detection setup
print("\n=== 2. Object Detection Framework ===")

class ObjectDetector(nn.Module):
    def __init__(self):
        super().__init__()
        # Backbone
        self.backbone = nn.Sequential(
            nn.Conv2d(3, 32, 3, padding=1),
            nn.ReLU(),
            nn.MaxPool2d(2, 2),
            nn.Conv2d(32, 64, 3, padding=1),
            nn.ReLU(),
            nn.MaxPool2d(2, 2),
        )

        # Bounding box regression
        self.bbox_head = nn.Sequential(
            nn.Linear(64 * 8 * 8, 128),
            nn.ReLU(),
            nn.Linear(128, 4)  # x, y, w, h
        )

        # Class prediction
        self.class_head = nn.Sequential(
            nn.Linear(64 * 8 * 8, 128),
            nn.ReLU(),
            nn.Linear(128, 10)  # 10 classes
        )

    def forward(self, x):
        features = self.backbone(x)
        features_flat = features.view(features.size(0), -1)

        bboxes = self.bbox_head(features_flat)
        classes = self.class_head(features_flat)

        return bboxes, classes

detector = ObjectDetector()
print(f"Detector parameters: {sum(p.numel() for p in detector.parameters()):,}")

# 3. Semantic Segmentation
print("\n=== 3. Semantic Segmentation U-Net ===")

class UNet(nn.Module):
    def __init__(self, num_classes=5):
        super().__init__()
        # Encoder
        self.enc1 = self._conv_block(3, 32)
        self.pool1 = nn.MaxPool2d(2, 2)
        self.enc2 = self._conv_block(32, 64)
        self.pool2 = nn.MaxPool2d(2, 2)

        # Bottleneck
        self.bottleneck = self._conv_block(64, 128)

        # Decoder
        self.upconv2 = nn.ConvTranspose2d(128, 64, 2, stride=2)
        self.dec2 = self._conv_block(128, 64)
        self.upconv1 = nn.ConvTranspose2d(64, 32, 2, stride=2)
        self.dec1 = self._conv_block(64, 32)

        # Final output
        self.out = nn.Conv2d(32, num_classes, 1)

    def _conv_block(self, in_channels, out_channels):
        return nn.Sequential(
            nn.Conv2d(in_channels, out_channels, 3, padding=1),
            nn.ReLU(inplace=True),
            nn.Conv2d(out_channels, out_channels, 3, padding=1),
            nn.ReLU(inplace=True)
        )

    def forward(self, x):
        enc1 = self.enc1(x)
        enc2 = self.enc2(self.pool1(enc1))
        bottleneck = self.bottleneck(self.pool2(enc2))

        dec2 = self.dec2(torch.cat([self.upconv2(bottleneck), enc2], 1))
        dec1 = self.dec1(torch.cat([self.upconv1(dec2), enc1], 1))

        return self.out(dec1)

unet = UNet(num_classes=5)
print(f"U-Net parameters: {sum(p.numel() for p in unet.parameters()):,}")

# 4. Transfer Learning
print("\n=== 4. Transfer Learning with Pre-trained Models ===")

try:
    # Load pre-trained ResNet18
    pretrained_model = models.resnet18(pretrained=True)
    num_ftrs = pretrained_model.fc.in_features
    pretrained_model.fc = nn.Linear(num_ftrs, 10)

    print(f"Pre-trained ResNet18 adapted for 10 classes")
    print(f"Parameters: {sum(p.numel() for p in pretrained_model.parameters()):,}")
except:
    print("Pre-trained models not available")

# 5. Image preprocessing and augmentation
print("\n=== 5. Image Preprocessing and Augmentation ===")

transform_basic = transforms.Compose([
    transforms.Resize((224, 224)),
    transforms.ToTensor(),
    transforms.Normalize(mean=[0.485, 0.456, 0.406],
                        std=[0.229, 0.224, 0.225])
])

transform_augmented = transforms.Compose([
    transforms.RandomRotation(20),
    transforms.RandomHorizontalFlip(),
    transforms.ColorJitter(brightness=0.2, contrast=0.2),
    transforms.RandomAffine(degrees=0, translate=(0.1, 0.1)),
    transforms.Resize((224, 224)),
    transforms.ToTensor(),
    transforms.Normalize(mean=[0.485, 0.456, 0.406],
                        std=[0.229, 0.224, 0.225])
])

print("Augmentation transforms defined")

# 6. Synthetic image data
print("\n=== 6. Synthetic Image Data Creation ===")

def create_synthetic_images(num_images=100, img_size=32):
    """Create synthetic images with shapes"""
    images = []
    labels = []

    for _ in range(num_images):
        img = np.ones((img_size, img_size, 3)) * 255

        # Randomly draw shapes
        shape_type = np.random.randint(0, 3)

        if shape_type == 0:  # Circle
            center = (np.random.randint(5, img_size-5), np.random.randint(5, img_size-5))
            radius = np.random.randint(3, 10)
            cv2.circle(img, center, radius, (0, 0, 0), -1)
            labels.append(0)

        elif shape_type == 1:  # Rectangle
            pt1 = (np.random.randint(0, img_size-10), np.random.randint(0, img_size-10))
            pt2 = (pt1[0] + np.random.randint(5, 15), pt1[1] + np.random.randint(5, 15))
            cv2.rectangle(img, pt1, pt2, (0, 0, 0), -1)
            labels.append(1)

        else:  # Triangle
            pts = np.array([[np.random.randint(0, img_size), np.random.randint(0, img_size)],
                           [np.random.randint(0, img_size), np.random.randint(0, img_size)],
                           [np.random.randint(0, img_size), np.random.randint(0, img_size)]])