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


Traditional neural networks fail when new images that are very similar to previously seen images that have been translated are fed as input to the model. Convolutional neural networks play a key role in addressing this shortcoming. This is enabled through the various mechanisms that are present in CNNs, including filters, strides, and pooling. Initially, we built a toy example to learn about how CNNs work. Then, we learned about how data augmentation helps in increasing the accuracy of the model by creating translated augmentations on top of the original image. After that, we learned about what different filters learn in the feature learning process so that we could implement a CNN to classify images.

Finally, we saw the impact that differing amounts of training data have on the accuracy of test data. Here, we saw that the more training data that is available, the better the accuracy of the test data. In the next chapter, we will learn about how to leverage various transfer learning...