Location:Training / Robot / Introduction to Computer Vision / Computational Complexity and Separability

This course is an introduction to computer vision, including the basic principles of image formation, camera imaging geometry, feature detection and matching, multi-view geometry including stereo, motion estimation and tracking, and classification. The course will cover basic development methods for relevant applications, including finding known models in images, recovering depth from stereo images, camera calibration, image stabilization, automatic alignment (e.g. panoramas), tracking and motion recognition.

This course focuses on cultivating learners' intuition and mathematical thinking, thereby allowing learners to understand the difference between theory and practice of problems. All algorithms are demonstrated in slides. But remember what Yogi Berra said: In theory, there is no difference between theory and difference, in practice, there is a difference. (Einstein said something similar). In this course, most of the time you won't need to apply high-level library functions, just low-to-mid-level algorithms to analyze images and extract structural information.