Michael Werman

Research Interests

• Computer Vision
• Geometric Algorithms
• Statistical Algorithms

Demos:

• Texture synthesis using statistical learning.
• Stabilization and Panorama using linear-programming

  ---

  ---

  
  

  Selected publications on-line

  Robust Real Fusing Time-of-Flight Depth and Color for Real-Time Segmentation and Tracking.

  A. Bleiweiss and M. Werman.
  Dynamic 3D Imaging 2009. 
  

  
  Fast and Robust Earth Mover's Distances.

  O. Pele and M. Werman.
  ICCV 2009. 
  Code
     

  
  A Linear Time Histogram Metric for Improved SIFT Matchings.

  O. Pele and M. Werman.
  ECCV 2008. 
  Code
  

  
  Robust Real Time Pattern Matching using Bayesian Sequential Hypothesis Testing

  O. Pele and M. Werman.
  PAMI, 2008. 
  Code
  

  
  Accelerating Pattern Matching or How Much Can You Slide?

  O. Pele and M. Werman.
  ACCV, 2007. 
  Code
  

  
  Vertical Parallax from Moving Shadows.

  Y. Caspi and M. Werman.
  CVPR, 2006. 
  

  
  The Bottleneck Geodesic: Computing Pixel Affinity.

  I. Omer and M. Werman.
  CVPR, 2006. 
  

  
  Affine Invariance Revisited.

  E. Begelfor and M. Werman.
  CVPR, 2006. 
  

  
  Image Specific Feature Similarities.

  I. Omer and M. Werman.
  ECCV, 2006. 
  

  
  The World is not (always) Flat or Learning Curved Manifolds.

  E. Begelfor and M. Werman.
  HUJI-CSE-LTR-2006-191  PAMI,  2006. 
  

  
  How to Put Probabilities on Homographies.

  E. Begelfor and M. Werman.
  IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, VOL. 27, NO. 10, OCTOBER 2005. 
  

  
  On using priors in affine matching.

  V. Govindu and M. Werman.
  Image and Vision Computing, V 22, 14, Dec 2004, Pages 1157-1164. 
  

  
  Using Natural Image Properties as Demosaicing Hints.

  I. Omer and M. Werman.
  ICCV 2004.
  

  
  Color Lines: Image Specific Color Representation.

  I. Omer and M. Werman.
  CVPR 2004.
  

  
  Simulation of Rain in Videos

  S. Starik and  M. Werman
  Texture03.
    
   rain videos  
  

  Unsupervised Clustering of Images using their Joint Segmentation

  Y. Seldin, S. Starik  and  M. Werman
  SCTV03.
      
  

  The Viewing Graph

  N. Levi and  M. Werman
  CVPR 2003, II:599-606.
      
  

  Study of Mutual Information in Perceptual Coding with Application for Low Bit-Rate Compression.

  A. Ben-Shalom, S. Dubnov and M. Werman 
  Fourth International Symposium on Independent Component Analysis and Blind Source Separation. ICA 2003 
  

  Improved Low bit-rate audio compression using reduced rank ICA instead of psychoacoustic modeling.

       A. Ben-Shalom, S. Dubnov and M. Werman 
       IEEE International Conference on Acoustics, Speech and Signal Processing. ICASSP2003 
  

  Fast Convolution

       M. Werman.
       WSCG 2003, Feb 2003.
      
  

  On using Priors in Affine Matching

       V. Govindu, and M. Werman.
       Indian Conference on Computer Vision, Graphics and Image Processing, 2002.    
  
      
  

  Gradient domain high dynamic range compression

       R. Fattal, D. Lischinski, and M. Werman.
       ACM Transactions on Graphics (Proc. ACM SIGGRAPH 2002), July 2002.
      
  

  OAll Points Considered: A Maximum Likelihood Method for Motion Recovery

       Daniel Keren and Ilan Shimshoni and  Liran Goshen and  Michael Werman.
       Theoretical Foundations of Computer Vision, Springer LNCS series 2616, 72-85, (2003).
  
      
  

  Parameter Estimates for a Pencil of Lines: Bounds and Estimators

       G. Speyer and M. Werman
       ECCV, 2002.
      
  

  A Bayesian Method for Fitting Parametric and Nonparametric Models to Noisy Data

       M. Werman and D. Keren
       PAMI, 23, 5, 528-534, 2001.
      
  

  Texture mixing and texture movie synthesis using statistical learning

       Z. Bar-Joseph, R. El-Yaniv, D. Lischinski, and  M. Werman
       IEEE Transactions on Visualization and Computer Graphics, 7(2), 2001, pp. 120-135
      
  

  Self Organization in Vision: Stochastic Clustering for Image Segmentation, Perceptual Grouping, and Image Database Organization

       Y. Gdalyahu, D. Weinshall and M. Werman
       IEEE Trans. on Pattern Analysis and Machine Intelligence, 23(10):1053-1074, 2001. 
      
  

  Structure from Motion using Points, Lines, and Intensities

       J. Oliensis and  M. Werman
       CVPR 2000, II:599-606.
      
  

  Model Based Pose Estimator using Linear Programming

       M. Ben-Ezra, S. Peleg and M. Werman
       ECCV2000
  

  Real-Time Motion Analysis with Linear Programming,

       M. Ben-Ezra, S. Peleg, and M. Werman
       Computer Vision and Image Understanding, Vol. 78, No. 1, Apr 2000, pp. 32-52.
      
  

  A Full Bayesian Approach to Curve and Surface Reconstruction

       D. Keren and M. Werman
       JMIV, 11, 27-43, 1999.
      
  

  Robot Localization using Uncalibrated Camera Invariants

       M. Werman, S. Banerjee, S. Dutta Roy and M. Qiu. 
       IEEE CVPR'99, II:353-359, June 23-25, Fort Collins, Colorado, 1999.
      
  

  Trajectory Triangulation over Conic Sections

       A. Shashua, S. Avidan and M. Werman
       International Conference on Computer Vision (ICCV)  330-336, Sep., 1999.  
      
  

  A method for on-line clustering of non-stationary data

       I.D. Guedalia, M. London and M Werman
       Neural Computation 11:551-571,  1999.
  

  Real-Time Object Tracking from a Moving Video Camera: A Software Approach on a PC

       Y. Rosenberg and M. Werman
       IEEE Workshop on Applications of Comuter Vision, Princeton, Oct 1998, pp. 238-239.  
      
  

  Representing local motion as a probability distribution matrix applied to object tracking

       Y. Rosenberg and M. Werman
       CVPR, 1997, pp. 654--659.  
      
  

  A General Filter for Measurements with any Probability Distribution

       Y. Rosenberg and M. Werman
       CVPR, 1997, pp. 106--111.  
      
  

  On View Likelihood and Stability

       D. Weinshall and M. Werman,
       IEEE Trans. on Pattern Analysis and Machine Intelligence, 19(2):97-108, 1997.
      
  

  Duality Of Multi-Point And Multi-Frame Geometry: Fundamental Shape Matrices And Tensors

       D. Weinshall, M. Werman and A. Shashua,
       ECCV-96, II:217-227,  Cambridge, April 1996.
      
  

  Similarity and Affine Invariant Distance Between Point Sets

       M. Werman and D. Weinshall
       PAMI,  17(8), pp. 810-814, August 1995.
  

  The Study of 3D-from-2D using Elimination

       M. Werman and A. Shashua
       Int. Conf. Computer Vision, Boston, June 1995, 473-479.
  

  Fitting a Second Degree Curve if Both Coordinates are Subject to Error

       M. Werman and Z. Geyzel
       PAMI, 17,207--211, 1995.
  

  Trilinearity of Three Perspective Views and its Associated Tensor

       A. Shashua and M. Werman
       Int. Conf. Computer Vision, Boston, June 1995, 920-925.
  

  Linear Time Euclidean Distance Transform and Voronoi Diagram Algorithms

       H. Breu and J. Gil and D. Kirkpatrck and M. Werman
       PAMI, 17, 529--533, 1995.
  

  Pose Estimation by Fusing Noisy Data of Different Dimensions

       Y. Hel-Or and M. Werman
       Published: T-PAMI Vol 17, No. 2, Feb 1995.
  

  Highlight and Reflection Independent Multiresolution Textures from Image Sequences

       E. Ofek, E. Shilat, A. Rappoport, and M. Werman
       IEEE Computer Graphics and Applications, 17, 18--29.
  

  Model Based Pose Estimation of Articulated and Constrained Objects

       Y. Hel-Or, M. Werman
       ECCV-94, Stockholm, 267--273, May 1994.
  

  Stability and Likelihood of Views of Three Dimensional Objects

       D. Weinshall, M. Werman and N. Tishby
       3-ECCV, Stockholm, May 1994, pp. 24--35.
  

  Constraint-Fusion for Interpretation of Articulated Objects

       Y. Hel-Or, M. Werman
       CVPR-94, Seattle, June 1994, pp. 39--45.
  

  Similarity and Affine Distance Between Point Sets

       M. Werman, D. Weinshall
       12-ICPR, Jerusalem,  Vol I, pp. 723-725, October 1994.
  

  Matching Points into Pairwise Disjoint Noise Regions: Combinatorial Bounds and Algorithms Subject to Error

       E. M. Arkin and K. Kedem and J. S. B. Mitchell and J. Sprinzak and M. Werman
       ORSA Journal on Computing, special issue on computational geometry, 27-52, 1992.
  

  Probabilistic Analysis of Regularization

       D. Keren and M. Werman
       PAMI, 15, 983-1001, 1993.
      
  

  Computing 2D Min, Max and Median Filters

       Y. Gil and  M. Werman
       PAMI, 15, 504-507, 1993.
  

  Finding the Repeated Median Regression Line

       A. Stein and M. Werman
       3'rd Symposium on Discrete Algorithms, 409--413, 1992.
  

  Robust Statistics in Shape Fitting

       A. Stein and M. Werman
       Computer Vision and Pattern Recognition, 540-546, 1992.
  

  Segmenting and Compressing by Minimal Length Encoding

       D. Keren and R. Marcus and M. Werman
       10'th International Conference on Pattern Recognition 681--683, 1990.
   

  A Unified Approach to the Change of Resolution: Space and Gray-Level

       S. Peleg and M. Werman and H. Rom
       PAMI 11, 739-742
  

  Inverting the autocorrelation and the problem of locating points on a line, given unlabelled distances between them

       P. Lemke and M. Werman
       Tech Report, 1988.
  

  Halftoning as Optimal Quantization

      S. Peleg and M. Werman
     8th ICPR, 1986, pp. 1114-1116.
  

  Bipartite graph matching for points on a line or a circle

    M. Werman, S. Peleg, R. Melter, and T.Y. Kong
     Journal of Algorithms, Vol. 7, 1986, pp. 277-284.
  

  A distance metric for multidimensional histograms

     M. Werman, S. Peleg, and A. Rosenfeld
     CVGIP, Vol. 32, Dec. 1985, pp. 328-336.
  

  Min max operators in texture analysis

     M. Werman and S. Peleg
     Trans. on PAMI, Vol 7, Nov. 1985, pp. 730-733.
  

  
  

  I am always looking to supervise bright and motivated graduate students in areas relating to computer vision, image processing, graphics and geometric or statistical algorithms. Please take at least a cursory look at some of my on line papers before contacting me.