• Home
  • Projects
  • Home
  • Projects

MAHDI RAD

Senior Scientist at Microsoft

About me

I am a senior scientist at Microsoft Mixed Reality & AI Lab in Zurich. Before that I was a postdoctoral scholar at the Institute of Computer Graphics and Vision of Graz University of Technology (TUG). I received my Ph.D. degree with distinction from TUG, under supervision of Prof. Vincent Lepetit, in 2018. Before that, I obtained my B.Sc. and M.Sc. degrees in Computer Science from EPFL, Switzerland, in 2012 and 2014. My research interests include machine learning methods for computer vision, specifically methods for 3D scene understanding, 3D object pose estimation, 3D hand pose estimation, hand-object interaction, semi supervised learning, and domain adaptation.

Contact

  Mahdi Rad 
   mahdirad@microsoft.com 
  
Google Scholar  
   
Microsoft Mixed Reality & AI, Zurich, Switzerland

Selected Projects

MonteFloor: Extending MCTS for Reconstructing Accurate Large-Scale Floor Plans

ICCV 2021, ORAL.

[PDF][ video1][Video2]

General 3D Room Layout from a Single View byRender-and-Compare

ECCV, 2020.

[PDF]

ALCN: Adaptive Local Contrast Normalization

CVIU, 2020.

[PDF][ video1][Video2]

HOnnotate: A method for 3D Annotation of Hand and Object Poses

CVPR 2020.

[PDF][​video]

Domain Transfer for 3D Pose Estimation fromColor Images without Manual Annotations

ACCV 2018, ORAL.

[PDF][ video]

Making Deep Heatmaps Robust to PartialOcclusions for 3D Object Pose Estimation

ECCV 2018.

[PDF][

Feature Mapping for Learning Fast and Accurate 3D Pose Inference from Synthetic Images

CVPR 2018.

[PDF][ video1][video2]

BB8: A Scalable, Accurate, Robust to Partial Occlusion Method for Predictingthe 3D Poses of Challenging Objects without Using Depth

ICCV 2017.

[PDF][ video1][video2][video3][video4]

Last update - Nov 2021.

Website Templates

created with

Website Builder Software .