Seeing : From Image analysis to 3D Computer vision

Catalogue des cours de Télécom SudParis

Code

IGSF IMA 4522

Niveau

M1

Graduate

Graduate

Semestre

Spring

Domaine

Image

Programme

Programme Ingénieur

Langue

Anglais/English

Crédits ECTS

4

Heures programmées

39

Charge de travail

90

Coordonnateur(s)

Département

  • Electronique et Physique

Equipe pédagogique

Organisation

Cours/TD/TP/projet/examen : 33/0/6/0

Acquis d'apprentissage

At the end of the course unit, students will be able to:
- explain and use algorithms for image pre-processing, segmentation and 2D analysis,
- model the view of a 3D scene and calibrate a camera,
- model and determine the geometry of a couple of cameras for stereovision,
- describe some advanced applications of embedded vision,
- explain algorithms for human detection and face and body perception by computer vision.

Compétences CDIO

  • 1.1.1 - Mathématiques (y compris statistiques)
  • 1.3 - Connaissances avancées en ingénierie : méthodes et outils
  • 2.1.2 - Modélisation
  • 3.3.1 - Communication en anglais

Mots-clés

- image analysis - computer vision - Image segmentation - Mathematical morphology - 2D image registration - Aspect graph - Hough Transform - Variational methods - projective models - projective invariants, - camera calibration, - stereovision, - 3D sensors. - 3D shape features.

Contenu

Lectures :
- Computer vision applications in industry
- Image segmentation: edges, regions, textures
- Mathematical morphology
- 2D image registration
- 2D shape features. Aspect graph
- Hough Transform
- Variational methods for image processing and computer vision
- 3D scene analysis: projective models & invariants, camera calibration, stereovision, epipolar geometry, fundamental matrix,autocalibration
- 3D sensors. 3D shape features. 3D/2D registration, model-based vision
- Vision systems

Labs :
- Image segmentation
- Mathematical morphology
- Determination of the epipolar geometry between two images of the same scene

Evaluation

Final grades will be based on lab work reports along the course unit

Formule de l'évaluation

Final mark = Average (Lab works).

Fiche mise à jour le 18/06/2018