This course provides a hands-on, project-driven approach to Computer Vision, using two powerful tools: OpenCV for image processing and PyTorch for building deep learning models. Learners begin with image fundamentals—pixels, color spaces, image histograms, and filtering techniques. Using OpenCV, students practice transformations like rotation, scaling, blurring, edge detection (Canny, Sobel), thresholding, contour detection, and morphological operations. The course then transitions to object detection and recognition using classical techniques like Haar cascades, HOG + SVM, and motion tracking. The deep learning component introduces convolutional neural networks (CNNs), explaining convolutional layers, pooling, and feature maps. Students build models from scratch in PyTorch and train them on datasets such as MNIST, CIFAR-10, and custom images. Advanced topics include transfer learning with pretrained models like ResNet, YOLO, and EfficientNet, as well as semantic segmentation with U-Net and instance segmentation using Mask R-CNN. Data augmentation, GPU acceleration, and hyperparameter tuning are covered in practical labs. Learners also deploy vision models to mobile apps or edge devices using TorchScript and ONNX. Real-world projects involve face detection, number plate recognition, medical imaging, and quality control in manufacturing. By course end, students are equipped to develop, train, and deploy vision models for applications across industries—from healthcare to retail, robotics to autonomous systems.