Computer Vision

A comprehensive and in-depth introduction to computer vision theories and deep learning technologies.

全面、深入介绍计算机视觉理论和深度学习技术

This book primarily introduces and explores a series of core topics in computer vision, including camera imaging, image processing, analysis and perception, and 3D reconstruction. It begins with an introduction to camera models, imaging processes, as well as color models, illumination models, and rendering models. Subsequently, it systematically covers various tasks such as image filtering, feature extraction, and image and video perception and understanding. For each task, the book not only provides a detailed explanation of classical methods but also comprehensively introduces cutting-edge deep learning-based approaches. Finally, it discusses the geometric principles involved in 3D reconstruction, the reconstruction pipeline, and novel deep learning-based methods.

Each core problem is addressed in its own chapter, with emphasis on explaining fundamental concepts. The book makes extensive use of color illustrations to help readers understand the material. It is suitable as a textbook for undergraduate and graduate courses in computer vision and digital image processing, and can also serve as a reference book for deep learning courses. Additionally, it provides a valuable resource for professionals and researchers interested in computer vision.

本书主要介绍和探讨计算机视觉的一系列核心主题，包括相机成像，图像处理、分析和感知，三维重建等。首先介绍了相机模型、成像过程以及图像的颜色模型、照射模型、渲染模型等；然后系统性地介绍了图像滤波、特征提取、图像和视频感知与理解等多种任务，不仅详细讲解了各任务中的经典方法，还全面地介绍了前沿的基于深度学习的方法；最后介绍了三维重建中涉及的几何原理、重建步骤以及基于深度学习的新方法。

本书针对每一个核心问题单独成章，并着重讲解基本概念。通过大量的彩图，帮助读者理解问题。适合作为本科及研究生的计算机视觉和数字图像处理课程的教材，并且可以作为深度学习课程的参考书。同时，也可供对计算机视觉感兴趣的相关专业人士参考。