Book Image

Modern Computer Vision with PyTorch

By : V Kishore Ayyadevara, Yeshwanth Reddy
5 (1)
Book Image

Modern Computer Vision with PyTorch

5 (1)
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

Why leverage neural networks for image analysis?

In traditional computer vision, we would create a few features for every image before using them as input. Let's take a look at a few such features based on the following sample image in order to appreciate the effort that we are avoiding going to by training a neural network:

Note that we will not walk you through how to get these features as the intention here is to help you realise why creating features manually is a sub-optimal exercise:

  • Histogram feature: For some tasks, such as auto-brightness or night vision, it is important to understand the illumination in the picture; that is, the fraction of pixels that are bright or dark. The following graph shows a histogram for the example image. It depicts that the image is well illuminated since there is a spike at 255:
  • Edges and Corners feature: For tasks such as image segmentation, where it is important to find the set of pixels corresponding to each person, it makes sense...