Dynamic Scene Analysis

Course Description

Fundamental concepts of computer vision. Extraction of dynamic information from image sequences. Image differencing. Background subtraction. Optical flow estimation methods. Motion-based image segmentation. Object detection and tracking. Using active contour models (snakes) for object tracking. Model-based object tracking. Probabilistic approaches to object tracking. Differential extraction and tracking of rectangular features (Harris, KLT). Matching invariant features (DOG, SIFT). Geometrical aspects of computer vision: image formation, camera calibration, geometry of one, two and multiple views. Robust parameter estimation. Applications: video surveilance, traffic surveilance, object motion analysis, video compression, navigation, traffic sign detection, recognition and tracking.

Study Programmes

Post-graduation study


David Forsyth, Jean Ponce (2003.), Computer Vision: A Modern Approach, Prentice Hall
Simon J. D. Prince (2012.), Computer Vision, Cambridge University Press
Richard Hartley, Andrew Zisserman (2004.), Multiple View Geometry in Computer Vision, Cambridge University Press
Yi Ma, Stefano Soatto, Jana Kosecka, S. Shankar Sastry (2004.), An Invitation to 3-D Vision, Springer
Emanuele Trucco, Alessandro Verri (1998.), Introductory Techniques for 3-D Computer Vision, Prentice Hall


ID 154684
  Winter semester
L0 English Level
L1 e-Learning