Description |
1 online resource (226 pages) 5 tables, 23 halftones and 37 line drawings |
Series |
Data-Enabled Engineering |
Contents |
Cover -- Half tilte -- Series title -- Title page -- Copyright -- Dedication -- Contents -- Preface -- Acknowledgments -- Authors -- Chapter 1: Introduction to Visual Computing -- Image Representation Basics -- Transform-Domain Representations -- Image Histograms -- Image Gradients and Edges -- Going beyond Image Gradients -- Line Detection Using the Hough Transform -- Harris Corners -- Scale-Invariant Feature Transform -- Histogram of Oriented Gradients -- Decision-Making in a Hand-Crafted Feature Space -- Bayesian Decision-Making |
|
Decision-Making with Linear Decision BoundariesA Case Study with Deformable Part Models -- Migration toward Neural Computer Vision -- Summary -- References -- Chapter 2: Learning As a Regression Problem -- Supervised Learning -- Linear Models -- Least Squares -- Maximum-Likelihood Interpretation -- Extension to Nonlinear Models -- Regularization -- Cross-Validation -- Gradient Descent -- Geometry of Regularization -- Nonconvex Error Surfaces -- Stochastic, Batch, and Online Gradient Descent -- Alternative Update Rules Using Adaptive Learning Rates |
|
MomentumSummary -- References -- Chapter 3: Artificial Neural Networks -- The Perceptron -- Multilayer Neural Networks -- The Back-Propagation Algorithm -- Improving BP- Based Learning -- Activation Functions -- Weight Pruning -- Batch Normalization -- Summary -- References -- Chapter 4: Convolutional Neural Networks -- Convolution and Pooling Layer -- Convolutional Neural Networks -- Summary -- References -- Chapter 5: Modern and Novel Usages of CNNs -- Pretrained Networks -- Generality and Transferability |
|
Using Pretrained Networks for Model CompressionMentee Networks and FitNets -- Application Using Pretrained Networks: Image Aesthetics Using CNNs -- Generative Networks -- Autoencoders -- Generative Adversarial Networks -- Summary -- References -- Appendix A:Yann -- Structure of Yann -- Quick Start with Yann: Logistic Regression -- Multilayer Neural Networks -- Convolutional Neural Network -- Autoencoder -- Summary -- References -- Postscript -- References |
Summary |
This book covers the fundamentals in designing and deploying techniques using deep architectures. It is intended to serve as a beginner's guide to engineers or students who want to have a quick start on learning and/or building deep learning systems. This book provides a good theoretical and practical understanding and a complete toolkit of basic information and knowledge required to understand and build convolutional neural networks (CNN) from scratch. The book focuses explicitly on convolutional neural networks, filtering out other material that co-occur in many deep learning books on CNN topics |
Bibliography |
Includes bibliographical references and index |
Subject |
Computer vision.
|
|
Neural networks (Computer science)
|
|
COMPUTERS -- Software Development & Engineering -- Systems Analysis & Design.
|
|
TECHNOLOGY & ENGINEERING -- Engineering (General)
|
|
Computer vision.
|
|
Neural networks (Computer science)
|
Form |
Electronic book
|
Author |
Li, Baoxin
|
ISBN |
9781498770408 |
|
1498770401 |
|
1315154285 |
|
9781315154282 |
|