机器视觉与机器学习

课程编码：1801010812I0P1006Y 英文名称：Computer vision and Machine Learning 课时：40 学分：2.00 课程属性：学科核心课 主讲教师：王伟强

This course focuses on the fundamental principles and methods of machine vision and machine learning. The course content includes image formation and visual principles, image filtering theory, color, texture, and shape feature description, edge detection and region segmentation, stereoscopic vision, pattern learning, object classification, deep learning theory, and its application in visual problems. Through the study of this course, students will gain a solid understanding of the basic theories and methods of computer vision and modern deep machine learning. They will also gain insight into the current research status and main methods in this field, providing them with a theoretical basis and practical skills for further research in this subject area.

大学基础数学（线性代数，高等数学，概率论）

第一章 Introduction and Course Review 2.0学时 王伟强
第1节 Introduction and Course Review
第二章 Geometric Model and Calibration of Camera 5.0学时 王伟强
第1节 image imaging ( pinhole perspective, weak perspective, camera with lens, human eyes)
第2节 internal and external parameters
第3节 geometric calibration of camera
第三章 Light and Shadow 4.0学时 王伟强
第1节 pixel brightness
第2节 shadow estimation
第3节 model mutual reflection
第4节 shape of a shadow image
第四章 Color 2.0学时 王伟强
第1节 human color perception
第2节 color physics
第3节 color representation
第4节 image color model
第5节 color based inference
第五章 Linear filtering 2.0学时 王伟强
第1节 linear filtering and convolution
第2节 shift-invariant linear systems
第3节 spatial frequency and Fourier transform
第4节 sampling and aliasing
第5节 filters and templates
第6节 technology: normalization of correlation and detection modes
第7节 technology: scale and image pyramid
第六章 Local image features 2.0学时 王伟强
第1节 calculate the image gradient
第2节 characterization of image gradient
第3节 find corner points and establish neighbors
第4节 neighbors are described through SIFT feature and HOG feature
第5节 actual calculation of local features
第七章 Texture 2.0学时 王伟强
第1节 local texture characterization using filters
第2节 texture characterization by pooling texture primitives
第3节 texture synthesis and filling of holes in the image
第4节 image denoising
第5节 restore shape from texture
第八章 Stereoscopic vision 4.0学时 王伟强
第1节 geometric properties and polar constraints of binocular cameras( polar geometry, eigenmatrix, basic matrix)
第2节 binocular reconstruction
第3节 human stereoscopic vision
第4节 local algorithm of binocular fusion
第5节 global algorithm of binocular fusion
第6节 use of multiple cameras
第7节 application: robot navigation
第九章 Neural networks 2.0学时 王伟强
第1节 overview of neural networks
第2节 back propagation network and BP learning algorithm
第3节 Big data and deep learning
第十章 clustering-based segmentation 4.0学时 王伟强
第1节 human vision: grouping and gestalt principles
第2节 important applications
第3节 image segmentation based on pixel clustering
第4节 segmentation, clustering and graph theory
第5节 application of image segmentation in practice
第十一章 Grouping and model fitting 4.0学时 王伟强
第1节 Hough transform and fit the line and plane
第2节 Robustness ( m-estimation method, RANSAC: search normal)
第3节 The probability model was used for fitting
第4节 Motion segmentation based on parameter estimation
第5节 model selection: which is the best
第十二章 Convolutional neural networks and semantic segmentation 3.0学时 王伟强
第1节 Convolutional neural network and tricks
第2节 Semantic segmentation and implementations of deep neural networks
第十三章 Object detection and tracking 4.0学时 王伟强
第1节 Convolutional neural networks and object detection
第2节 detection based tracking
第3节 match tracking
第4节 linear dynamic model tracking based on kalman filter
第5节 particle filtering

Wang Weiqiang is currently a professor and doctoral supervisor at the School of Computer Science and Technology, University of Chinese Academy of Sciences. His main research interests include computer vision, machine learning, and intelligent human-computer interaction. He joined the Institute of Computing Technology, Chinese Academy of Sciences in May 2001 as an assistant researcher and was promoted to associate researcher in October 2001. In August 2003, he joined the School of Information Science and Technology, Graduate University of Chinese Academy of Sciences as an associate professor. From October 2004 to September 2005, he was a visiting scholar at Carnegie Mellon University in the United States. He was appointed as a professor in June 2009. He has undertaken research and development work for key projects such as the 863 Program, the National Science and Technology Support Program sub-project, the National Natural Science Foundation of China, key international cooperation and exchange projects, and key research and development projects. He has published more than 130 research papers in important domestic and international journals and conferences.