Per citar aquest document:
Visual Recognition [43088]
Vanrell Martorell, Maria
Baldrich, Ramón
Valveny, Ernest
Benavente, Robert
Giró, Xavier
Rossinyol, Marçal
Fornés, Alicia
Lladós, Josep
Gonzàlez, Jordi
van de Weije, Joost
Universitat Autònoma de Barcelona. Escola d'Enginyeria

Data: 2015-16
Resum: In Computer Vision, visual recognition corresponds to the task of explaining the content of an image in terms of “What is there?”. This question can be addressed in many different ways, and in this Module 5 we will consider two particular types of answers: the category of the object appearing in an image, and the identification of a particular object instance. In other words: looking for cars vs. looking for “my car”. The contents will span from the Bag of Words pipeline for automatic category classification to the use of efficient large-scale techniques for instance retrieval. This module will develop the most recent and popular techniques in Computer Vision which can answer the aforementioned questions. The project to be developed during this Module will provide the students with the practical insights to the theoretical content of the module.
Llengua: Anglès.
Pla d'estudis: Visió per Computador / Computer Vision [1172]

3 p, 137.0 KB

El registre apareix a les col·leccions:
Materials de curs > Programes de curs (Guies docents)

 Registre creat el 2016-02-09, darrera modificació el 2016-02-11

   Favorit i Compartir