Cognitive Vision Systems

Recommended Prerequisites

Algebra, Differential Calculus, Probability.

Teaching Methods

Teaching methods include lectures by the professor, presentations of specific topics by the students and also tutorial supervision.

Learning Outcomes

The goals and outcomes of this unit include learn how to extract information from images so that it is possible to estimate 3D world and object structure, object motion (including velocities and displacements), object recognition, shapes and activities. The techniques that the student has to learn are based on learning and classification.

Work Placement(s)

No

Syllabus

Introduction to probability, Fitting probability models, Learning and inference in vision, Regression and Classification models, Graphical Models, Models for chains, trees and grids, Models for shape, style and identity, Models for visual words.

Head Lecturer(s)

Hélder de Jesus Araújo

Assessment Methods

Assessment
Research work: 100.0%

Bibliography

"Computer Vision: Models, learning and inference",  Simon Prince

"Cognitive Vision Systems: Sampling the Spectrum of Approaches" (Lecture Notes in Computer Science), Henrik I. Christensen and Hans-Hellmut Nagel .

"The Cognitive Neuroscience of Vision" (Fundamentals of Cognitive Neuroscience), Martha J. Farah

"Active Vision: The Psychology of Looking and Seeing" (Oxford Psychology Series), John M. Findlay and Iain D. Gilchrist.

"Pattern Recognition and Machine Learning", Christopher M. Bishop

"Learning with Kernels", Bernhard Schlkopf and Alexander J. Smola