Lê-Huu, Đ.Khuê

Doing research in optimization, machine learning, computer vision.

I am currently a postdoctoral researcher at the THOTH team of Inria Grenoble Rhône-Alpes, working with Karteek Alahari since June 2020. From April 2018 to February 2020, I led the AI team of Qopius, a computer vision startup based in Paris (acquired by Trax in February 2020). Prior to that, I obtained a PhD in Computer Science from CentraleSupélec, Université Paris-Sacaly, where I was advised by Nikos Paragios.

news

Oct 26, 2021 Initial release of CRF, a library for dense CRFs. Stay tuned for our semantic segmentation code.
Sep 28, 2021 Our work on regularized Frank-Wolfe for MAP inference has been accepted at NeurIPS 2021.
Jun 1, 2020 I joined Inria’s THOTH team to work with Karteek Alahari as a postdoctoral researcher.

selected publications

  1. NeurIPS
    Regularized Frank-Wolfe for Dense CRFs: Generalizing Mean Field and Beyond
    Đ.Khuê Lê-Huu, Karteek Alahari
    In Advances in Neural Information Processing Systems 2021
  2. CVPR
    Continuous Relaxation of MAP Inference: A Nonconvex Perspective
    Đ.Khuê Lê-Huu, Nikos Paragios
    In IEEE Conference on Computer Vision and Pattern Recognition 2018
  3. CVPR
    Alternating Direction Graph Matching
    Đ.Khuê Lê-Huu, Nikos Paragios
    In IEEE Conference on Computer Vision and Pattern Recognition 2017