This concise introduction covers the core concepts and methods of computer vision. Chapter 1 introduces image representation learning—from traditional modeling to deep learning–based approaches—and explores 2D, 3D, and video vision from a deep learning perspective. Chapters 2–3 address low-level tasks (e.g., super-resolution, denoising); Chapters 4–6 focus on mid-level tasks (e.g., feature matching, 3D reconstruction, depth estimation); and Chapters 7–11 examine high-level tasks (e.g., classification, semantic segmentation, object detection, metric learning) alongside classic theories. Chapters 12–13 discuss advances such as diffusion models, model compression, foundation models, and vision-language models.

Readers will gain practical skills for research and real-world applications. This book is ideal as a textbook for courses in computer science, automation, and electronic information engineering, as well as a self-study reference for practitioners.

LU Jiwen, Tenured Associate Professor, Ph.D. Supervisor, and Associate Chair of the Department of Automation at Tsinghua University. He is a recipient of the National Science Fund for Distinguished Young Scholars, an IEEE Fellow, an IAPR Fellow, and Editor-in-Chief of Pattern Recognition Letters. His research spans computer vision, pattern recognition, and intelligent robotics. He has authored over 160 papers in PAMI, CVPR, ICCV, and ECCV, and holds more than 50 licensed national invention patents.