Book Image

Modern Computer Vision with PyTorch

By : V Kishore Ayyadevara, Yeshwanth Reddy
Book Image

Modern Computer Vision with PyTorch

By: V Kishore Ayyadevara, Yeshwanth Reddy

Overview of this book

Deep learning is the driving force behind many recent advances in various computer vision (CV) applications. This book takes a hands-on approach to help you to solve over 50 CV problems using PyTorch1.x on real-world datasets. You’ll start by building a neural network (NN) from scratch using NumPy and PyTorch and discover best practices for tweaking its hyperparameters. You’ll then perform image classification using convolutional neural networks and transfer learning and understand how they work. As you progress, you’ll implement multiple use cases of 2D and 3D multi-object detection, segmentation, human-pose-estimation by learning about the R-CNN family, SSD, YOLO, U-Net architectures, and the Detectron2 platform. The book will also guide you in performing facial expression swapping, generating new faces, and manipulating facial expressions as you explore autoencoders and modern generative adversarial networks. You’ll learn how to combine CV with NLP techniques, such as LSTM and transformer, and RL techniques, such as Deep Q-learning, to implement OCR, image captioning, object detection, and a self-driving car agent. Finally, you'll move your NN model to production on the AWS Cloud. By the end of this book, you’ll be able to leverage modern NN architectures to solve over 50 real-world CV problems confidently.
Table of Contents (25 chapters)
Section 1 - Fundamentals of Deep Learning for Computer Vision
Section 2 - Object Classification and Detection
Section 3 - Image Manipulation
Section 4 - Combining Computer Vision with Other Techniques


In this chapter, we understood the need for a single network that performs both feature extraction and classification in a single shot, before we learned about the architecture and the various components of an artificial neural network. Next, we learned about how to connect the various layers of a network before implementing feedforward propagation to calculate the loss value corresponding to the current weights of the network. We next implemented backpropagation to learn about the way to optimize weights to minimize the loss value. Further, we learned about how the learning rate plays a role in achieving optimal weights for a network. In addition, we implemented all the components of a network – feedforward propagation, activation functions, loss functions, the chain rule, and gradient descent to update weights in NumPy from scratch so that we have a solid foundation to build upon in the next chapters.

Now that we understand how a neural network works, we'll implement one using PyTorch in the next chapter, and dive deep into the various other components (hyperparameters) that can be tweaked in a neural network in the third chapter.