José Lezama

IIE, Facultad de Ingenieria, Universidad de la República
Julio Herrera y Reissig 565
Montevideo 11300

jlezama at fing edu uy

Short Bio

I am an Assistant Professor at Universidad de la República, Uruguay. Before I worked as a postdoc with Prof. Pablo Musé. Before that I was a postdoc at Duke University working with Guillermo Sapiro and Sina Farsiu. I did my Phd with Prof. Jean-Michel Morel at CMLA, ENS Cachan (France) and Prof. Gregory Randall at IIE, Universidad de la República (Uruguay) and with the wonderful collaboration of Rafael Grompone von Gioi. Previous to that I worked at INRIA/ENS with Josef Sivic and Ivan Laptev. Before all that I graduated as Electrical Engineer from Universidad de la República, Uruguay.
My research interests are in Computer Vision and Machine Intelligence in general. Lately I've been focused on learning useful image representations with geometric constraints.


Teaching / Seminars

Deep Learning for Artificial Vision (2020)
Estimation and Prediction in Time Series (2020)
CS Seminar, Universidad de Chile (January 2020)
EE Seminar, Universidade Federal do Rio de Janeiro (December 2019)
Mathematics of Imaging, Institut Henri Poincaré (March 2019)
Deep Reinforcement Learning (February 2019)
Deep Learning For Artificial Vision (2018, assistant)
Reinforcement Learning and AlphaGo (May 2018)
Deep Learning For Artificial Vision (2017, assistant)

Selected Publications

Overcoming the Disentanglement vs Reconstruction Trade-off via Jacobian Supervision
José Lezama
ICLR 2019 (Article) | Github | Video
OLÉ: Orthogonal Low-rankEmbedding, A Plug and Play Geometric Loss for Deep Learning
José Lezama, Qiang Qiu, Pablo Musé, Guillermo Sapiro
IEEE CVPR 2018 (Article) | Github
Not Afraid of the Dark: NIR-VIS Face Recognition via Cross-spectral Hallucination and Low-rank Embedding
José Lezama*, Qiang Qiu*, Guillermo Sapiro
In IEEE CVPR 2017 (Article) | Project page with Source Code
Segmentation guided registration of wide field-of-view retinal optical coherence tomography volumes
José Lezama, Dibyendu Mukherjee, Ryan P. McNabb, Guillermo Sapiro, Anthony N. Kuo, and Sina Farsiu
In Biomedical Optics Express 2016 (Article) | Project page with Source Code
A Contrario Detection of Good Continuation of Points
José Lezama, Rafael Grompone von Gioi, Gregory Randall and Jean-Michel Morel
In IEEE ICIP 2014 (PDF) | Vision Research (2015) (link) | Online Demo (user: demo, password: demo)
Finding Vanishing Points via Point Alignments in Image Primal and Dual Domains
José Lezama, Rafael Grompone von Gioi, Gregory Randall and Jean-Michel Morel
In IEEE CVPR 2014 | PDF | Project Page | MATLAB code
A Contrario 2D Point Alignment Detection
José Lezama, Jean-Michel Morel, Gregory Randall and Rafael Grompone von Gioi
In IEEE TPAMI (2014) | PDF | Online Demo | MATLAB code
Track to the future: Spatio-temporal video segmentation with long-range motion cues
Jose Lezama, Karteek Alahari, Josef Sivic and Ivan Laptev
In IEEE CVPR 2011 | PDF | Project page

Full list of publications in Google Scholar

My Github