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

Introducing RNNs

An RNN can have multiple architectures. Some of the possible ways of architecting an RNN are as follows:

In the preceding diagram, the boxes at the bottom are the input, followed by the hidden layer (the middle boxes), and then the boxes at the top are the output layer. The one-to-one architecture is a typical neural network with a hidden layer between the input and output layers. Examples of different architectures are as follows:

  • One-to-many: The input is an image and the output is a caption of the image.
  • Many-to-one: The input is a movie review (multiple words in input) and the output is the sentiment associated with the review.
  • Many-to-many: Machine translation of a sentence in one language to a sentence in another language.

The idea behind the need for RNN architecture

RNNs are useful when we want to predict the next event given a sequence of events. An example of that could be to predict the word that comes after this: This is an ___.

Let's say that in reality...