Convolutional Neural Networks in Visual Computing

Designing and deploying deep architectures can feel overwhelming at first, and this volume aims to give newcomers a manageable entry point. It walks through the core principles needed to conceive, train and put convolutional neural networks to work, with the goal of enabling readers to construct these systems themselves.

The approach balances conceptual grounding with hands-on practice. Readers receive both the theory behind how these networks operate and a practical set of basic tools and knowledge, so that they can follow the logic of a CNN and then build one step by step.

The book is aimed at engineers and students who want a fast route into deep learning. It concentrates solely on convolutional networks and leaves aside the wider range of topics that usually appear alongside them in general deep learning texts.