Dr. Àgata Lapedriza
Staff member
webpage || alapedriza(at)uoc.edu
Research topic/area:
Computer Vision. Image Understanding.
Currently working on:
Scene Understanding.
Areas of Research:
Object Detection, Object Recognition, Scene Classification, Scene Understanding, Facial Expression Analysis.

Agata Lapedriza received her MS degree in Mathematics at the Universitat de Barcelona (UB), and her PhD degree in Computer Science at the Computer Vision Center (CVC), at the Universitat Autònoma Barcelona (UAB). Currently she is an Associate Professor at the Universitat Oberta de Catalunya (UOC).

As a researcher, she is mainly interested in Scene Perception and Understanding. From September 2012 to June 2015 she worked as a visiting researcher at the Computer Science and Artificial Intelligence Laboratory (CSAIL) of the Massachusetts Institute of Technology (MIT). At MIT CSAIL she was involved in a project focused on the use of deep Convolutional Neural Networks (CNN) for scene recognition. The project involved several stages, such as the creation of a big dataset of scene categories, called Places Database, and the training of a CNN using this dataset. The deep CNN trained with Places Database is currently the state-of-the-art in scene categorization. With the same team of researchers, Agata Lapedriza also worked on understanding the internal representations learned by the CNNs and showed that several of the units, particularly in the upper layers, learned to detect objects (with no supervision).

During her PhD she was working on Multitask Learning (Machine Learning) applied to face classification. Currently she is still working on facial expression analysis at the UOC. In particular her more recent work in face analysis presents a system for detecting subtle facial emotions, such as positiveness or negativeness, using Action Units as mid-level features.

At this moment all of her teaching activities rely on academic direction and coordination of MS degree’s subjects and programs. She has a wide experience in advising BS and MS negree projects in the context of computer vision and artificial intelligence, and she is currently advising 2 PhD students.

In terms of affiliations, she is a researcher of the CVC and also a member of the BCN Perceptual Computing Lab research group (UB-UOC). Furthermore, she has open collaborations with two research groups at the MIT CSAIL department.


Journal Articles


Sanchez-Mendoza, David; Masip, David; Lapedriza, Àgata

Emotion recognition from mid-level features (Journal Article)

Pattern Recognition Letters, 67 , pp. 66–74, 2015.

(BibTeX | Links: )


Igual, Laura; Lapedriza, Àgata; Borràs, Ricard

Robust gait-based gender classification using depth cameras (Journal Article)

EURASIP J. Image and Video Processing, 2013 , pp. 1, 2013.

(BibTeX | Links: )


Masip, David; Lapedriza, Àgata; Vitrià, Jordi

Boosted Online Learning for Face Recognition (Journal Article)

IEEE Transactions on Systems Man and Cybernetics part B, 39 , pp. 530–538, 2009.


Caballé, Santi; Lapedriza, Àgata; Masip, David; Xhafa, Fatos; Abraham, Ajith

Enabling Automatic Just-in-time Evaluation of In-class Discussions in On-line Collaborative Learning Practices. (Journal Article)

Journal of Digital Information Management (JDIM), 7 , pp. 290-297, 2009.



Keil, Matthias; Lapedriza, Agata; Masip, David; Vitria, Jordi

Preferred Spatial Frequencies for Human Face Processing Are Associated with Optimal Class Discrimination in the Machine (Journal Article)

PLoS ONE, 3 (7), pp. e2590, 2008.

(BibTeX | Links: )

Lapedriza, Àgata; Seguí, Santi; Masip, David; Vitrià, Jordi

A sparse Bayesian approach for joint feature selection and classifier learning (Journal Article)

Pattern Anal. Appl., 11 (3-4), pp. 299–308, 2008, ISSN: 1433-7541.

(BibTeX | Links: )


Lapedriza, Agata; Masip, David; Vitrià, Jordi

On the Use of External Face Features for Identity Verification. (Journal Article)

JOURNAL OF MULTIMEDIA, 1 (4), pp. 11-20, 2006.




Zhou, Bolei; Lapedriza, Agata; Xiao, Jianxiong; Torralba, Antonio; Oliva, Aude

Learning deep features for scene recognition using places database (Inproceeding)

Advances in neural information processing systems, pp. 487–495, 2014.