Diane Larlus

  • machine learning
  • deep learning
  • computer vision
  • data analytics
  • artificial intelligence

I am a senior research scientist in the Computer Vision group. My research interests include machine learning applied to computer vision. I am particulary interested in visual scene understanding, and related tasks such as visual search, the joint localization of objects and their parts, and reasoning at the scene-level with images and text.

Short Bio

I obtained a M.Sc. in Image, Vision and Robotics from UJF/INP, Grenoble, France, in 2005. From 2005 to 2008, I worked as a doctoral candidate in the LEAR group, at INRIA Grenoble. During the summer 2007, I did an internship at the JRL/AIST laboratory in Tsukuba, Japan. I obtained my Ph.D. in 2008, from INP Grenoble. From 2008 to 2010, I worked as a post-doc at TU Darmstadt, Germany. I joined what has now become Naver Labs Europe in 2010.


  • 2 papers accepted at CVPR17:
    • AnchorNet: a weakly supervised network to learn geometry-sensitive features for semantic matching.
    • Beyond instance-level image retrieval: Leveraging human captions to learn representations for semantic visual search.
  • ECCV 16 best reviewer award
  • 1 paper accepted at ECCV16:  Deep Image Retrieval: Learning global representations for image search


Publications since I joined Xerox are available here: XRCE Publications

Full list available on DBLP or Google Scholar.

External collaborations

Ongoing projects:

  • Svetlana Lazebnik, from the University of Illinois at Urbana-Champaign (UAC)
  • Andrea Vedaldi, from the University of Oxford (joint super vision of David Novotny)

Previous external collaborations

  • Sabine Süsstrunk, from EPFL, around scene understanding using visible and near-infrared data (UAC, then Open Innovation project, 2010 to 2013)
  • Andrew Zisserman and Andrea Vedaldi, from Oxford, around object recognition (UAC, 2012-2015)
+33 (0)4 76 61 50 67