John Oliensis ( CV )
Associate Professor
Department of Computer Science
Stevens Institute of Technology
Castle Point on Hudson
Hoboken, NJ 07030

email: (my last name) (at)


CS 558 Computer Vision

CS 541 Artificial Intelligence

CS 601 Complexity

CS 559 Machine Learning

Analysis of Algorithms

Edge-based shape tracking results

Simple template matching using the shape similarity metric described in Hongzhi Wang and JO, Rigid Shape Matching by Segmentation Averaging
(PAMI, to appear). No motion or background information was used (no background subtraction)

PETS 2007 indoor surveillance (fixed camera): occlusion, lighting change, cluttered background similar to target, pose change, some appearance change (articulation), some scale change.
PETS 2007: 1 ,   PETS 2007: 2 ,   PETS 2007: 3 ,   PETS 2007: 4

Sequences from David Ross
(compression artifacts cause some jitter---see zipped full results)

trellis70 ( zipped uncompressed )     outdoor, moving camera and target, lighting change, pose change, some scale change   
Dudek   ( zipped uncompressed )   moving camera and target, occlusion, scale change, pose change, appearance change, lighting change, indoor
davidin300 ( zipped uncompressed )     indoor, moving camera and target, lighting change, pose change, some scale and appearance change
Ming-Hsuan Yang ( zipped uncompressed )    indoor, fixed camera, scale change, lighting change, occlusion
Plush toy (aka Sylvester) ( zipped uncompressed )     Large pose change, scale change, indoor, fixed camera
Car 11 ( zipped uncompressed )     outdoor, night, moving camera and target, lighting change and glare, some scale change
Car 4 ( zipped uncompressed )     outdoor, moving camera and target, lighting change, scale change
fish ( zipped uncompressed )     lighting change, moving camera, fixed object

For comparisons to the WSL of Jepson et al, Mean Shift, and the algorithm of Ross et al, see (from David Ross ):
3-fold compare: trellis70 ,   3-fold compare: Dudek ,   3-fold compare: Sylvester (plush toy) , 3-fold compare: davidin300 ,
In these comparisons, highlighted ellipse = WSL (Jepson et al), green dotted rectangle = Mean Shift, yellow rectangle = Ross, Lim, and M-H Yang.
You may want to install this VLC viewer to see these comparison sequences.

Edge-preserving image smoothing: results

Based on the similarity metric described in Hongzhi Wang and JO, Rigid Shape Matching by Segmentation Averaging
(PAMI, to appear). The smoothing technique is non-local: it is sensitive to the large scale image structure

Movie 1 ,   Movie 2

Segmentation results: simple greedy merging

Based on the similarity metric described in Hongzhi Wang and JO, Rigid Shape Matching by Segmentation Averaging
(PAMI, to appear). See the paper for more results and quantitative comparisons to other segmentation algorithms.

Movie 1 ,   Movie 2 ,   Movie 3 ,   Movie 4

NECI Vision Workshops

Recent Papers

  • Rigid Shape Matching by Segmentation Averaging (PAMI) and ECCV 2008 , Hongzhi Wang and JO ( initial paper , Nov 2006)
  • Generalizing Edge Detection to Contour Detection for Image Segmentation , abstract , Hongzhi Wang and JO
    A condensed version of this paper appeared in POCV Workshop 2006.
  • Iterative extensions of the Sturm/Triggs algorithm: convergence and nonconvergence , abstract , PAMI 29:12 2217-2233, 2007.
    @2007 IEEE. Personal use of this material is permitted. However, permission to reprint/republish this material for advertising or promotional purposes or for creating new collective works
    for resale or redistribution to servers or lists, or to reuse any copyrighted component of this work in other works must be obtained from the IEEE.

  • J. Oliensis, The Least-Squares Error for Structure from Infinitesimal Motion , Abstract IJCV Vol. 61, No. 3, 259--299, 2005, and ECCV04 (short version) © 2004 by Springer-Verlag .
  • Shan Lu, D. Metaxas, D. Samaras, J. Oliensis, ``Using multiple cues for hand tracking and model refinement,'' CVPR II 443-450, 2003.
  • R. Vidal and J. Oliensis, Structure from planar motions with small baselines, Abstract, ECCV II: 383-398, 2002 © 2002 by Springer-Verlag

  • J. Oliensis, Linear Stratified Self-Calibration and Euclidean Reconstruction, Abstract, NECI TR, 2002. A condensed version of this paper appeared as part of Fast and Accurate Self-Calibration in ICCV 745-752, 1999, © 1999 by IEEE

  • J. Oliensis, Exact Two-Image Structure from Motion , Abstract, PAMI 24:12, 1618-1633, 2002. © 2002 by IEEE Also, see NECI TR 2001 , the older but longer version. These papers extend the previous paper: ``New Algorithms for Two-Frame Structure from Motion," ICCV 737-744, 1999.

  • J. Oliensis and Y. Genc, Fast and Accurate Algorithms for Projective Multi-Image Structure from Motion ( (pdf) ), Abstract, PAMI 23:6 546-559, 2001. © 2001 by IEEE This extends the previous paper in ICCV 536-542 1999.
  • J. Oliensis, Direct Multi-Frame Structure from Motion for Hand-Held Cameras, Abstract, ICPR I 889-895, 2000. © 1999 by IEEE

  • J. Oliensis and Mike Werman, Structure from Motion using Points, Lines, and Intensities ( pdf), Abstract, CVPR II 599-606, 2000. © 1999 by IEEE

  • J. Oliensis, Fast and Accurate Self-Calibration, ( (pdf)), Abstract, ICCV 745-752, 1999. © 1999 by IEEE

  • J. Oliensis, A Multi-frame Structure from Motion Algorithm under Perspective Projection, Abstract,
    IJCV 34:2/3 163-192, 1999. Revised version of previous April 1997 TR. © 1999 by Kluwer academic

  • J. Oliensis, A New Structure From Motion Ambiguity, Abstract,
    PAMI 22:7, 685-700, 2000 and CVPR 185-191, 1999. © 1999 by IEEE

  • J. Oliensis, Recovering Heading and Structure for Constant-Direction Motion, Abstract,
    NECI Technical Report, October 1997, revised April 2000. A condensed version of this paper appears as `Computing the Camera Heading from Multiple Frames,'' in CVPR 203-210, 1998.

  • J. Oliensis, A Critique of Structure from Motion Algorithms, pdf , Abstract,
    CVIU 80 172-214,2000. (This version corrects misprints in the published discussion paper.) © 2000 by ACADEMIC PRESS

  • J. Oliensis and V. Govindu, An Experimental Study of Projective Structure from Motion, Abstract,
    PAMI 21:7 665-671, 1999. © 1999 by IEEE

  • J. Oliensis, Structure from Linear or Planar Motions,
    CVPR 335-342, 1996. © 1995 by IEEE

  • J. Oliensis, Rigorous Bounds for Two-Frame Structure from Motion,
    NECI Technical Report, October 1995. A condensed version of this report appears in ECCV I 184-195, 1996.

  • J. Oliensis, Multiframe Structure from Motion in Perspective,
    NECI Technical Report, April 1995. A slightly condensed version of this report appeared in the Workshop on Representations of Visual Scenes, Boston, June 1995, 77-84. © 1995 by IEEE

  • J. Oliensis, A Linear Solution for Multiframe Structure from Motion,
    Proc. Image Understanding Workshop, Monterey, California, November 1994, pp. 1225-1231.

    Other Publications

    Shape from Shading

  • Original 1991 paper, simultaneous with Rouy et al.
    J. Oliensis and P. Dupuis, Direct method for reconstructing shape from shading , Proc. SPIE Conf. 1570 on Geometric Methods in Computer Vision, pp. 116-128, 1991.

    This paper gives the first description of the optimal control approach to shape from shading. It presents a modern algorithm similar to that proposed by Rouy and Tourin, SIAM J. Numerical Analysis 29:3, 867-884, 1992. Unlike in Rouy/Tourin, the algorithm does not require traditional-sense boundary conditions and works for light from any direction. We note in this paper that the algorithm can be extended to discontinuous intensity functions and nondifferentiable surfaces.

  • P. Dupuis and J. Oliensis, An Optimal Control Formulation and Related Numerical Methods for a Problem in Shape Reconstruction , Annals of Applied Probability , Vol. 4, No. 2, 1994, pp. 287-346.

    The above paper gives the full presentation of our shape-from-shading work, including algorithmic details and correctness proofs. Since the paper is long, I have also divided it into the following 2 pdf files: Optimal Control...First half and Optimal Control...Second half

  • J. Oliensis and P. Dupuis, ``Direct Method for Reconstructing Shape from Shading,'' in Physics-Based Vision: Principles and Practice, Shape Inference Volume, L. Wolff, S. Shafer, G. Healey, editors, Jones and Bartlett, Boston, June 1992, pp. 17-28.

    This paper introduces small improvements in our 1991 algorithm that speed it up dramatically. In general, we find that 4-8 passes over the image suffice to reconstruct the surface.

  • J. Oliensis, Shape from Shading as a Partially Well-Constrained Problem , Computer Vision, Graphics, and Image Processing: Image Understanding , Vol. 54, No. 2, 1991, pp. 163-183.

  • Paul Dupuis and J. Oliensis, ``Shape from Shading: Provably Convergent Algorithms and Uniqueness Results,'' in {\it Fourth European Conference on Computer Vision}, 1994, Vol. 2, pp. 259-268.

  • J. Oliensis and P. Dupuis, ``A Global Algorithm for Shape from Shading,'' long paper, {\it Proc. of the Fourth International Conference on Computer Vision}, Berlin, Germany 1993, pp. 692-701.

  • J. Oliensis, Uniqueness in Shape From Shading , International Journal of Computer Vision , Vol. 6 No. 2, pp. 75-104, 1991.

    Recovery of Affine Deformations

  • R. Manmatha and J. Oliensis, ``Measuring the Affine Transform - I: Recovering Scale and Rotation,'' {\it IEEE Computer Vision and Pattern Recognition}, New York, NY 1993, pp. 754-755. Also, technical report CMPSCI TR 92-74, University of Massachusetts at Amherst, 1992.

  • R. Manmatha and J. Oliensis, ``Measuring the Affine Transform - I: Recovering Scale and Rotation,'' in {\it Proc. DARPA Image Understanding Workshop}, San Diego, California, 1993, pp. 449-458.


  • J. Thomas and J. Oliensis, Dealing with Noise in Multi-Frame Structure from Motion, CVIU Vol 76., No. 2, pp. 109-124, 1999.

  • J. I. Thomas, A. Hanson, J. Oliensis, ``Refining 3D Reconstructions: A Theoretical and Experimental Study of the Effect of Cross-Correlations,'' Computer Vision, Graphics, and Image Processing: Image Understanding, Vol. 60, 1994 pp. 359-370.

  • H. Sawhney, J. Oliensis, and A. Hanson, ``Image Description and 3D Reconstruction from Image Trajectories of Rotational Motion,'' IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol. 15, No. 9, 1993, pp. 885-898.


  • J. Oliensis, ``Local Reproducible Smoothing Without Shrinkage,'' IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol. 15, No. 3, 1993, pp. 307-311.


    First Workshop: FEB. 27 - MAR. 10, 1995

    Princeton, N.J.

    Organizer: John Oliensis


    The NECI Vision Workshop brought together vision psychologists and computer vision researchers for a two week period to exchange ideas. The meeting was oriented toward discussion with a relaxed schedule of presentations. Foci of discussion included object recognition, recovery of structure and motion, subjective contours, perceptual inference, and low level vision.


    Bill Bialek (NECI), Heinrich Bulthoff (Max-Planck), Brian Burns (Teleos), Jacob Feldman (Rutgers), Ingemar Cox (NECI), David Forsyth (Berkeley), Jonas Garding (KTH, Sweden), Richard Hartley (GE), David Jacobs (NECI), Allan Jepson (U. of Toronto), Dan Kersten, (U. of Minnesota), David Knill (U. of Pennsylvania), Tony Lindeberg (KTH, Sweden), Mike Langer (McGill), Zili Liu (NECI), Larry Maloney (NYU), Steve Maybank (GEC/U. of Oxford), John Oliensis (NECI), Pietro Perona (Cal Tech), Jean Ponce (U. of Illinois), Harpreet Sawhney (IBM), Bob Shapley (NYU), Stefano Soatto (Cal Tech), Mike Tarr (Yale), Shimon Ullman (Weizmann), Bill Warren (Brown), Lance Williams (NECI), Alan Yuille (Harvard), Steve Zucker (McGill).
  • 2nd NEC Research Institute Vision Workshop (1996)


    Tao Alter (MIT), Jonas August (McGill), Bart Anderson (MIT), Ronen Basri (Weizmann Institute), Peter Belhumeur (Yale), Irving Biederman (USC), Peter Blicher (NECI), Patrick Cavanagh (Harvard), Ingemar Cox (NECI), Sven Dickinson (Rutgers), James Elder (NECI/York), Davi Geiger (NYU), Alan Gilchrist (Rutgers), Mel Goodale (U. of Western Ontario), Keith Humphrey (U. of Western Ontario), Glyn Humphreys (U. of Birmingham), Dave Jacobs (NECI), Phil Kellman (UCLA), David Kriegman (Yale), Ilona Kovacs (Rutgers), Michael Langer (NECI), Zili Liu (NECI), David Lowe (U. of British Columbia), Pascal Mamassian (New York Univ), John Oliensis (NECI), Steve Omohundro (NECI), Thomas Papathomas (Rutgers), V.S. Ramachandran (UC San Diego), Ron Rensink (Nissan), Ruth Rosenholtz (NASA Ames), Nava Rubin (Harvard/NYU), Kaleem Siddiqui (McGill/Yale), Eric Saund (Xerox PARC), Pawan Sinha (MIT), Rudiger von der Heydt (Johns Hopkins), Joachim Weickert (Utrecht U. Hospital), Laurie Wilcox (Universite de Montreal/York), Lance Williams (NECI), Steve Zucker (McGill/Yale)

  • Third NEC Research Institute Vision Workshop (1998)

    Lucretia Moroni + Fatto a Mano

  • New Fatto a Mano Home Page
  • New York Times Article Dec 9 1999
  • Kids Art New York: Free children's art classes (apart from supplies) at KidsArt, Fatto-a-mano
  • Adult designer workshops for professional, artists, designers.

  • Previous Life: The NEC Research Institute

    NECI Vision Group

  • Past and upcoming seminars