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.

Preface

Who this book is for

What this book covers

To get the most out of this book

Section 1 - Fundamentals of Deep Learning for Computer Vision

Section 1 - Fundamentals of Deep Learning for Computer Vision

Free Chapter

Artificial Neural Network Fundamentals

Artificial Neural Network Fundamentals

Comparing AI and traditional machine learning

Learning about the artificial neural network building blocks

Implementing feedforward propagation

Implementing backpropagation

Putting feedforward propagation and backpropagation together

Understanding the impact of the learning rate

Summarizing the training process of a neural network

PyTorch Fundamentals

PyTorch Fundamentals

Installing PyTorch

PyTorch tensors

Building a neural network using PyTorch

Using a sequential method to build a neural network

Saving and loading a PyTorch model

Building a Deep Neural Network with PyTorch

Building a Deep Neural Network with PyTorch

Representing an image

Why leverage neural networks for image analysis?

Preparing our data for image classification

Training a neural network

Scaling a dataset to improve model accuracy

Understanding the impact of varying the batch size

Understanding the impact of varying the loss optimizer

Understanding the impact of varying the learning rate

Understanding the impact of learning rate annealing

Building a deeper neural network

Understanding the impact of batch normalization

The concept of overfitting

Section 2 - Object Classification and Detection

Section 2 - Object Classification and Detection

Introducing Convolutional Neural Networks

Introducing Convolutional Neural Networks

The problem with traditional deep neural networks

Building blocks of a CNN

Implementing a CNN

Classifying images using deep CNNs

Implementing data augmentation

Visualizing the outcome of feature learning

Building a CNN for classifying real-world images

Transfer Learning for Image Classification

Transfer Learning for Image Classification

Introducing transfer learning

Understanding VGG16 architecture

Understanding ResNet architecture

Implementing facial key point detection

Multi-task learning – Implementing age estimation and gender classification

Introducing the torch_snippets library

Practical Aspects of Image Classification

Practical Aspects of Image Classification

Generating CAMs

Understanding the impact of data augmentation and batch normalization

Practical aspects to take care of during model implementation

Basics of Object Detection

Basics of Object Detection

Introducing object detection

Creating a bounding box ground truth for training

Understanding region proposals

Understanding IoU

Non-max suppression

Mean average precision

Training R-CNN-based custom object detectors

Training Fast R-CNN-based custom object detectors

Advanced Object Detection

Advanced Object Detection

Components of modern object detection algorithms

Training Faster R-CNN on a custom dataset

Working details of YOLO

Training YOLO on a custom dataset

Working details of SSD

Training SSD on a custom dataset

Test your understanding

Image Segmentation

Image Segmentation

Exploring the U-Net architecture

Implementing semantic segmentation using U-Net

Exploring the Mask R-CNN architecture

Implementing instance segmentation using Mask R-CNN

Applications of Object Detection and Segmentation

Applications of Object Detection and Segmentation

Multi-object instance segmentation

Human pose detection

Image colorization

3D object detection with point clouds

Section 3 - Image Manipulation

Section 3 - Image Manipulation

Autoencoders and Image Manipulation

Autoencoders and Image Manipulation

Understanding autoencoders

Understanding convolutional autoencoders

Understanding variational autoencoders

Performing an adversarial attack on images

Performing neural style transfer

Generating deep fakes

Image Generation Using GANs

Image Generation Using GANs

Introducing GANs

Using GANs to generate handwritten digits

Using DCGANs to generate face images

Implementing conditional GANs

Advanced GANs to Manipulate Images

Advanced GANs to Manipulate Images

Leveraging the Pix2Pix GAN

Leveraging CycleGAN

Leveraging StyleGAN on custom images

Super-resolution GAN

Section 4 - Combining Computer Vision with Other Techniques

Section 4 - Combining Computer Vision with Other Techniques

Training with Minimal Data Points

Training with Minimal Data Points

Implementing zero-shot learning

Implementing few-shot learning

Combining Computer Vision and NLP Techniques

Combining Computer Vision and NLP Techniques

Introducing RNNs

Introducing LSTM architecture

Implementing image captioning

Transcribing handwritten images

Object detection using DETR

Combining Computer Vision and Reinforcement Learning

Combining Computer Vision and Reinforcement Learning

Learning the basics of reinforcement learning

Implementing Q-learning

Implementing deep Q-learning

Implementing deep Q-learning with the fixed targets model

Implementing an agent to perform autonomous driving

Moving a Model to Production

Moving a Model to Production

Understanding the basics of an API

Creating an API and making predictions on a local server

Moving the API to the cloud

Using OpenCV Utilities for Image Analysis

Using OpenCV Utilities for Image Analysis

Drawing bounding boxes around words in an image

Detecting lanes in an image of a road

Detecting objects based on color

Building a panoramic view of images

Detecting the number plate of a car

Other Books You May Enjoy

Other Books You May Enjoy

Leave a review - let other readers know what you think

Appendix

Chapter 1 - Artificial Neural Network Fundamentals

Chapter 2 - PyTorch Fundamentals

Chapter 3 - Building a Deep Neural Network with PyTorch

Chapter 4 - Introducing Convolutional Neural Networks

Chapter 5 - Transfer Learning for Image Classification

Chapter 6 - Practical Aspects of Image Classification

Chapter 7 - Basics of Object Detection

Chapter 8 - Advanced Object Detection

Chapter 9 - Image Segmentation

Chapter 11 - Autoencoders and Image Manipulation

Chapter 12 - Image Generation Using GANs

Chapter 13 - Advanced GANs to Manipulate Images

Chapter 14 - Training with Minimal Data Points

Chapter 15 - Combining Computer Vision and NLP Techniques

Chapter 16 - Combining Computer Vision and Reinforcement Learning

Customer Reviews

5 star

0

4 star

0

3 star

0

2 star

0

1 star

0

3D object detection with point clouds

So far, we have learned how to predict a bounding rectangle on 2D images using algorithms that have the core underlying concept of anchor boxes. We will now learn how the same concept can be extended to predict 3D bounding boxes around objects.

In a self-driving car, tasks such as pedestrian/obstacle detection and route planning cannot happen without knowing the environment. Predicting 3D object locations along with their orientations becomes an important task. Not only is the 2D bounding box around obstacles important, but also knowing the distance from the object, height, width, and orientation of the obstacle are critical to navigating safely in the 3D world.

In this section, we will learn how YOLO is used to predict the 3D orientation and position of cars and pedestrians on a real-world dataset.

The instructions for downloading the data, training, and testing sets are all given in this GitHub repo: https://github.com/sizhky/Complex-YOLOv4-Pytorch...