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

Implementing image captioning

Image captioning means generating a caption given an image. In this section, we will first learn about the preprocessing to be done to build an LSTM that can generate a text caption given an image, and then will learn how to combine a CNN and LSTM to perform image captioning. Before we learn about building a system that generates captions, let's understand how a sample input and output might look:

In the preceding example, the image is the input and the expected output is the caption of the image – In this image I can see few candles. The background is in black color.

The strategy that we will adopt to solve this problem is as follows:

  1. Preprocess the output (ground truth annotations/captions) so that each unique word is represented by a unique ID.
  2. Given that the output sentences can be of any length, let's assign a start and end token so that the model knows when to stop generating predictions. Furthermore, ensure that all input sentences...