Martial Hebert

Professor
The Robotics Institute
School of Computer Science
Carnegie Mellon University
Pittsburgh, PA 15213, USA

Office: 224 EDSH
Phone: +1 (412) 268-2585
Email: hebert@ri.cmu.edu



Research

My work is in the areas of computer vision and perception for autonomous systems. My interests are in the interpretation of perception data (both 2-D and 3-D), including building models of environments. Current research directions include:
  • Efficient techniques for object/category recognition
  • Use of contextual information, in particular 3-D geometry from images, for scene analysis
  • Symbolic knowledge for scene interpretation and reconstruction
  • Motion analysis for feature extraction and event detection in video clips
  • Efficient tools for the analysis of dynamic 3-D point clouds ("3-D signal processing")
  • Perception for autonomous systems
  • Detection, tracking, and prediction in dynamic environments

Graduate Students

Post-Docs/Visitors/Research Staff


  • Christoph Mertz  (with C. Thorpe)
  • Soonyong Park
  • Luis Navarro-Serment   (with C. Thorpe)
  • Arne Suppe   (with C. Thorpe)

  • Ranqi Zhu 

  • Associated Projects/Labs/Centers

    Quality of Life Technology Center (QoLT)
    Vision and Mobile Robotics Lab
    Willow Research Team (INRIA)
    ASDMCon Project (Civil Eng. Dept.)

    Former Post-Docs/Visitors (Since 2008)

    Former Students (in approximate chronological order)

    • Marius Leordeanu (Robotics, PhD 2009)
    • PhD Thesis -- Spectral Graph Matching, Learning, and Inference for Computer Vision
    • Caroline Pantofaru  (Robotics, PhD 2008, now at Willow Garage)
      • PhD Thesis -- Studies In Using Image Segmentation to Improve Object Recognition
    • Ranjith Unnikrishnan (Robotics, PhD 2008, now at Google)
      • PhD Thesis -- Statistical Approaches to Multi-scale Point Cloud Processing
    • Andrew Stein (Robotics, PhD 2008, now at Tandent Vision)
      • PhD Thesis -- Occlusion Boundaries: Low-Level Processing to High-Level Reasoning

    Data/Software  

    Selected Projects (in no particular order, someday will be udpated...)

    (see the RI Web page for a mostly complete list of publications)
       Object discovery

    Hongwen Kang, Takeo Kanade, Martial Hebert

    aa Blocks World Revisited

    Abhinav Gupta, Alexei A. Efros, Martial Hebert
    Stacked Hierarchical Labeling
    Daniel Munoz, Drew Bagnell
    in ECCV 2010


    Structure Discovery in Multi-modal Data
    Alvaro Collet Romea, Siddhartha Srinivasa, Martial Hebert
    aa Addressing Feature Ambiguity for Point-Based 3D Object Recognition
    Edward Hsiao,  Alvaro Collet
    aa
    Contextual Classification with Functional Max-Margin Markov Networks
    Daniel Munoz,  Nicolas Vandapel, Drew Bagnell
    in CVPR'09
    AA Activity Recognition from First Person Sensing
    Ekaterina H. (Spriggs) Taralova, Fernando De la Torre, Martial Hebert
    aa Efficient image indexing for localization and image composition
    Takeo Kanade, Alexei A. Efros, Martial Hebert
    in 3dRR'09, EgoVision'09
    aa 3D Scene Analysis
    David Lee, Takeo Kanade, Martial Hebert
    in CVPR 2009
    ..
    An Empirical Study of Context in Object Detection
    Santosh Kumar Divvala, Derek Hoiem, James Hays, Alexei A. Efros, Martial Hebert
    in CVPR 2009
    localtion
    Image Matching in Large Scale Environments
    Hongwen Kang, Takeo Kanade, Alexei A. Efros, Martial Hebert
    event detection Volumetric Features for Event Recognition in Video
    Yan Ke, Rahul Sukhtankar, Martial Hebert
    in ICCV 2007.
    chair couch Detecting Boundaries for Segmentation and Recognition
    Andrew Stein, Martial Hebert
    in ICCV 2007.
    A framework for learning to recognize and segment object classes
    Caroline Pantofaru, Martial Hebert
    in BMVC 2007
    Object and Category Recognition from Pairwise Relations between Features
    Marius Leordeanu Rahul Sukhtankar, Martial Hebert
    in CVPR 2007

    Evaluation of Image Segmentation Algorithms
    Caroline Pantofaru, Ranjith Unnikrishnan, Martial Hebert
    in PAMI 2007.
    Recovering Occlusion Boundaries from a Single Image
    Derek Hoiem, Andrew Stein, Alexei A. Efros, Martial Hebert
    in ICCV 2007.
    Source code available
    Putting Objects in Perspective
    Derek Hoiem, Alexei A. Efros, Martial Hebert
    In CVPR 2006.
    Best Paper Award
    Source code available
    Geometric Context from a Single Image
    Derek Hoiem, Alexei A. Efros, Martial Hebert
    In ICCV 2005
     (see also expanded journal version, IJCV 2007)
    Executable available (for non-commerical use only)
    Spectral Techniques for Matching and Inference
    Marius Leordeanu,
    Martial Hebert
    in ICML 2006.
    Automatic Photo Pop-up
    Derek Hoiem, Alexei A. Efros, Martial Hebert
    In SIGGRAPH 2005
    See article in The Economist
    Technology licensed to FreeWebs as Fotowoosh
    3-D Recognition and 3-D environment model building


    terrain terrain2  Terrain Classification from 3-D Point Clouds for Autonomous Mobility
    Jean-François Lalonde,  Nicolas Vandapel, Martial Hebert
    In Journal of Field Robotics, 2006.
      Detection and Tracking of Moving Objects  in Dynamic Environments
    Christoph Mertz,  Luis Navarro-Serment,  Martial Hebert

    graph 3d Efficient Data Structures for On-Line Processing of 3-D Point Clouds
    Jean-François Lalonde,  Nicolas Vandapel, Martial Hebert
    In International Journal of Robotics Research, 2007.
    Wire Lalonde
    Automatic Scale Selection for Analyzing 3-D Point Clouds
    Ranjith Unnikrishnan, Jean-François Lalonde,  Nicolas Vandapel, Martial Hebert
    in BMVC 2007.

    Sponsors

    Misc.