Jakob Verbeek

I am a senior researcher at Inria Grenoble working on machine learning and computer vision in the Thoth research team.

Want to join the lab?

Open applications are welcome by outstanding candidates for postdoc positions in architecture learning for deep nets, unsupervised deep learning, semantic segmentation, and image captioning, or related topics

Other information

Project page for convolutional neural fabrics

My publications on Google Scholar , arXiv , HAL , DBLP.

Some recent papers

Publications HAL de verbeek de cordelia,schmid;jakob,verbeek;matthijs,douze;zaid,harchaoui;thomas,mensink;remi,ronfard;laurent,zwald;frederic,jurie;roger,mohr du labo/EPI 2398;44942

titre: Sparse weakly supervised models for object localization in road environment
auteur: Valentina Zadrija, Josip Krapac, Sini?a ?egvi?, Jakob Verbeek
article: Computer Vision and Image Understanding, Elsevier, 2018, pp.1-13. 〈10.1016/j.cviu.2018.10.004〉
Accès au texte intégral et bibtex

titre: Learning Disentangled Representations with Reference-Based Variational Autoencoders
auteur: Adrià Ruiz, Oriol Martinez, Xavier Binefa, Jakob Verbeek
article: 2018
Accès au texte intégral et bibtex

titre: Predicting Future Instance Segmentation by Forecasting Convolutional Features
auteur: Pauline Luc, Camille Couprie, Yann Lecun, Jakob Verbeek
article: ECCV 2018 - European Conference on Computer Vision, Sep 2018, Munich, Germany. pp.1-21
Accès au texte intégral et bibtex

titre: Coverage and quality driven training of generative image models
auteur: Konstantin Shmelkov, Thomas Lucas, Karteek Alahari, Cordelia Schmid, Jakob Verbeek
article: 2018
Accès au texte intégral et bibtex