Estudia

• Identificar, valorar i utilitzar tècniques i mètodes actuals i novedosos per a la resolució de projectes tecnològics i de recerca en la visió per computador i robòtica, l’enginyeria de control i sistemes intel•ligents.
• Poder fer front a una carrera de recerca en vistes a l'elaboració d'una tesis doctoral en un context altament tecnològic, o per a l'exercici professional en activitats de desenvolupament tecnològic, innovació i recerca.
• Interatuar amb èxit en entorns de treball internacionals i multidisciplinars.

• Learn the techniques of modelling and calibrating computer vision systems.
• Learn how 3D information of the real world can be computed from 2D image projections
• Learn the principles of stereovision, triangulation and pattern projection
• Learn some feature detecting and feature matching algorithms and how to remove false data associations

1. Introduction

          1.1. Course organization: Objectives, Overview, Contents, Bibliography, Evaluation, Practical Sessions

2. Camera Modelling

          2.1. The pinhole camera

          2.2. Intrinsic and extrinsic parameters

3. Camera Calibration

          3.1. Computing the calibration matrix

          3.2. Accuracy Evaluation

4. Image Primitives

          4.1. Interest point detectors

          4.2. Harris and Hessian detectors

          4.3. Similarity measures: SAD, SSD, Correlation

          4.4. Introduction to Scale invariant features

5. Planar Transformations

          5.1. Review of SIFT

          5.2. A hierarchy of transformations: Euclidean, Similarity, Affine, Projective

          5.3. Computing the homography matrix

6. Outlier Rejection

          6.1. Probabilistic methods

          6.2. Least Median of Squares

          6.3. Random Sampling Consensus

          6.4. Applications: Planar motion estimation, Mosaicing, etc.

7. Reconstruction from 2 views

          7.1. The principle of Triangulation

          7.2. Epipolar geometry

          7.3. Computing the Fundamental matrix

          7.4. Accuracy Evaluation

          7.5. Experimental Results

8. Cameras in the real world

          8.1. Comercial cameras

          8.2. Camera characteristics

9. Pattern Projection Techniques

          9.1. The principle of codification

          9.2. State of art

          9.3. Time multiplexing, spatial codification, direct codification

          9.4. Steps to implement a pattern projection technique

• Hartley, Richard (2003). Multiple view geometry in computer vision (2nd ed.). Cambridge [etc.]: Cambridge University Press.
• Ma, Yi (2004). An Invitation to 3-D vision : from images to geometric models. New York [etc.]: Springer, cop..

Activitats d'avaluació:

30 % Final Exam (Exam mark >= 4/10)
70 % Practical Sessions (Every exercise Mark >= 5/10)

To pass the module, the global mark must be >= 5/10