I have completed my PhD at INRIA Willow, advised by Josef Sivic, Ivan Laptev, and Cordelia Schmid. I graduated from Ecole polytechnique in 2016 and got a Masters Degree in Mathematics, Machine Learning and Computer Vision (MVA). My research interests include : understanding visual scenes, learning from few or weak data, joint modeling of images and language. More details can be found in my resume.

Publications

Detecting unseen visual relations using analogies
Julia Peyre, Ivan Laptev, Cordelia Schmid and Josef Sivic.
In Proc. ICCV 2019
Weakly-supervised learning of visual relations
Julia Peyre, Ivan Laptev, Cordelia Schmid and Josef Sivic.
In Proc. ICCV 2017 (Oral, acceptance rate 2.1%).

PhD Thesis

Learning to detect visual relations
Julia Peyre

Invited talks

June 2019 : Université Aix Marseille, Département Informatique et Interactions.
August 2018 : BNPP PRAIRIE Summer School.

Teaching

2016-2017 : Teaching Assistant, Introduction to Probabilistic Graphical Models, MVA, ENS Cachan.