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

Transcribing handwritten images

In the previous section, we learned about generating sequences of words from an input image. In this section, we will learn about generating sequences of characters with the image as input. Furthermore, we will learn about the CTC loss function, which helps in transcribing handwritten images.

Before we learn about the CTC loss function, let's understand the reason why the architecture that we saw in the image captioning section might not apply in handwritten transcription. Unlike in image captioning, where there is no straightforward correlation between the content in the image and the output words, in a handwritten image, there is a direct correlation between the sequence of characters present in the image and the sequence of output. Thus, we will follow a different architecture from what we designed in the previous section.

In addition, assume a scenario where an image is divided into 20 portions (assuming a scenario of a maximum of 20 characters...