A comprehensive, self-contained guide to how machines learn to see — to turn raw pixels into understanding: what objects are present, where they ar…

Welcome to Computer Vision: From Pixels to Perception.

A comprehensive, self-contained guide to how machines learn to see — to turn raw pixels into understanding: what objects are present, where they are, how a scene is laid out in three dimensions, and what is happening over time. This is the seventh volume in a series; it blends intuition, mathematics, and runnable code, and is the discriminative counterpart to the companion Image Generation book: where that book asks "how do I create an image?", this one asks "how do I understand one?"

This title is part of the ShriIra library and is free to read in full, right here — our small contribution to making world-class knowledge easy to reach.

A note on reading it: open the Contents menu at the top of the reader to jump between chapters, use the Aa menu to set a comfortable text size, theme (light, sepia, or night), and single- or two-page layout. Your place is saved automatically, so you can always pick up where you left off.

We hope it serves you well.

— Shriira Press

Contents

1. Chapter 1 — What Is Computer Vision?
2. Chapter 2 — Images, Pixels, and Formation
3. Chapter 3 — Classical Computer Vision
4. Chapter 4 — Convolutional Neural Networks for Vision
5. Chapter 5 — Image Classification and Architectures
6. Chapter 6 — Vision Transformers and Modern Backbones
7. Chapter 7 — Object Detection
8. Chapter 8 — Semantic and Instance Segmentation
9. Chapter 9 — Beyond 2D: Pose, Depth, Motion, and 3D
10. Chapter 10 — Self-Supervised and Representation Learning
11. Chapter 11 — Video Understanding
12. Chapter 12 — Datasets, Evaluation, and Deployment
13. Chapter 13 — Ethics: Facial Recognition, Surveillance, and Bias
14. Appendix A — Notation and Symbols
15. Appendix B — Further Reading