Academia.eduAcademia.edu

Outline

Title
Abstract
Introduction
Results
References
All Topics
Computer Science
Database Systems

Toward true 3D object recognition

Profile image of Jean PonceJean Ponce

2004

Abstract

This paper addresses the problem of recognizing three-dimensional (3D) objects in photographs and image sequences. It revisits viewpoint invariants as a local representation of shape and appearance, and proposes a unified framework for object recognition where object models consist of a collection of small (planar) patches, their invariants, and a description of their 3D spatial relationship. This approach is applied to two fundamental instances of the 3D object recognition problem: (1) modeling rigid 3D objects from a small set of unregistered pictures and recognizing them in cluttered photographs taken from unconstrained viewpoints, and (2) recognizing non-uniform texture patterns despite appearance variations due to non-rigid transformations and changes in viewpoint. It is validated through several experiments, and extensions to the analysis of video sequences and the recognition of object categories are briefly discussed.

Related papers

Object Recognition with 3D Models.
Bernd Heisele

We propose techniques for designing and training of pose-invariant object recognition systems using realistic 3d computer graphics models. We look at the relation between the size of the training set and the classification accuracy for a basic recognition task and provide a method for estimating the degree of difficulty of detecting an object. We show how to sample, align, and cluster images of objects on the view sphere. We address the problem of training on large, highly redundant data and propose a novel active learning method which generates compact training sets and compact classifiers.

View PDFchevron_right
3D object recognition based on a viewpoint-analysis
Nicola Ancona

Image and Vision Computing, 1992

View PDFchevron_right
3d object modeling and recognition using local a ne - invariant image descriptors and multi - view spatial constraints
Jean Ponce

2005

This article introduces a novel representation for three-dimensional (3D) objects in terms of local affine-invariant descriptors of their images and the spatial relationships between the corresponding surface patches. Geometric constraints associated with different views of the same patches under affine projection are combined with a normalized representation of their appearance to guide matching and reconstruction, allowing the acquisition of true 3D affine and Euclidean models from multiple unregistered images, as well as their recognition in photographs taken from arbitrary viewpoints. The proposed approach does not require a separate segmentation stage, and it is applicable to highly cluttered scenes. Modeling and recognition results are presented.

View PDFchevron_right
Recognition of 3D Object from One Image Based on Projective and Permutative Invariants
Francisco Sanchez

Lecture Notes in Computer Science, 2004

We present a new approach to recognize 3D objects in an image taking the available information in it. To reach this aim a method based on a two step algorithm will be developed. In the first step, each 3D model will be converted in a set of 2D images using an aspect graph algorithm in which the concept of visual event is defined as the moment when the projective invariants got from the model have changed. In the second step, we use these invariants to recognize the object situated in the scene. The main contribution to this work is the combination of two typical techniques used to object recognition like are aspect graph and projective invariants. We have obtained better results with this combinations than by using them separately.

View PDFchevron_right
Multi-scale free-form 3D object recognition using 3D models
peter yuen

Image and Vision Computing, 2001

The recognition of free-form 3D objects using 3D models under different viewing conditions based on the geometric hashing algorithm and global veri®cation is presented. The matching stage of the algorithm uses the hash-table prepared in the off-line stage. Given a scene of feature points, one tries to match the measurements taken at scene points to those memorised in the hash-table. The technique used for feature recovery is the generalisation of the CSS method (IEEE Trans. Pattern Anal. Mach. Intell., 14 (1992) 789±805), which is a powerful shape descriptor expected to be an MPEG-7 standard. Smoothing is used to remove noise and reduce the number of feature points to add to the ef®ciency and robustness of the system. The local maxima of Gaussian and mean curvatures are selected as feature points. Furthermore, the torsion maxima of the zero-crossing contours of Gaussian and mean curvatures are also selected as feature points. Recognition results are demonstrated for rotated and scaled as well as partially occluded objects. In order to verify match, 3D translation, rotation and scaling parameters are used for veri®cation and results indicate that our technique is invariant to those transformations. Our technique for smoothing and feature extraction is more suitable than level set methods or volumetric diffusion for object recognition applications since it is applicable to incomplete surface data that arise during occlusion. It is also more ef®cient and allows for accurate estimation of curvature values. q

View PDFchevron_right
View-Based Models of 3D Object Recognition and Class-Specific Invariance
TOMASO POGGIO

1994

This paper describes the main features of a view-based model of object recognition. The model tries to capture general properties to be expected in a biological architecture for object recognition. The basic module is a regularization network in which each of the hidden units is broadly tuned to a speci c view of the object to be recognized. The network output, which m a y be largely view independent, is rst described in terms of some simple simulations. The following re nements and details of the basic module are then discussed: (1) some of the units may represent only components of views of the object { the optimal stimulus for the unit, its \center", is e ectively a complex feature (2) the units' properties are consistent with the usual description of cortical neurons as tuned to multidimensional optimal stimuli (3) in learning to recognize new objects, preexisting centers may be used and modi ed, but also new centers may be created incrementally so as to provide maximal invariance (4) modules are part of a hierarchical structure: the output of a network may be used as one of the inputs to another, in this way synthesizing increasingly complex features and templates (5) in several recognition tasks, in particular at the basic level, a single center using view-invariant features may be su cient. Modules of this type can deal with recognition of speci c objects, for instance a speci c face under various transformations such as those due to viewpoint and illumination, provided that a su cient n umber of example views of the speci c object are available. An architecture for 3D object recognition, however, must cope { to some extent { e v en when only a single model view is given. The main contribution of this paper is an outline of a recognition architecture that deals with objects of a nice class undergoing a broad spectrum of transformations { due to illumination, pose, expression and so on { by exploiting prototypical examples. A nice class of objects is a set of objects with su ciently similar transformation properties under speci c transformations, such as viewpoint transformations. For nice object classes, we discuss two possibilities: (a) class-speci c transformations are to be applied to a single model image to generate additional virtual example views, thus allowing some degree of generalization beyond what a single model view could otherwise provide (b) class speci c, view-invariant features are learned from examples of the class and used with the novel model image, without an explicit generation of virtual examples.

View PDFchevron_right
3D Object Recognition Based On Constrained 2D Views
Levente Toth

1998

The aim of the present work was to build a novel 3D object recognition system capable of classifying man-made and natural objects based on single 2D views. The approach to this problem has been one motivated by recent theories on biological vision and multiresolution analysis. The project's objectives were the implementation of a system that is able to deal with simple 3D scenes and constitutes an engineering solution to the problem of 3D object recognition, allowing the proposed recognition system to operate in a practically acceptable time frame. The developed system takes further the work on automatic classification of marine phytoplank-(ons, carried out at the Centre for Intelligent Systems, University of Plymouth. The thesis discusses the main theoretical issues that prompted the fundamental system design options. The principles and the implementation of the coarse data channels used in the system are described. A new multiresolution representation of 2D views is presented, which provides the classifier module of the system with coarse-coded descriptions of the scale-space distribution of potentially interesting features. A multiresolution analysis-based mechanism is proposed, which directs the system's attention towards potentially salient features. Unsupervised similarity-based feature grouping is introduced, which is used in coarse data channels to yield feature signatures that are not spatially coherent and provide the classifier module with salient descriptions of object views. A simple texture descriptor is described, which is based on properties of a special wavelet transform. The system has been tested on computer-generated and natural image data sets, in conditions where the inter-object similarity was monitored and quantitatively assessed by human subjects, or the analysed objects were very similar and their discrimination constituted a difficult task even for human experts. The validity of the above described approaches has been proven. The studies conducted with various statistical and artificial neural network-based classifiers have shown that the system is able to perform well in all of the above mentioned situations. These investigations also made possible to take further and generalise a number of important conclusions drawn during previous work carried out in the field of 2D shape (plankton) recognition, regarding the behaviour of multiple coarse data channels-based pattern recognition systems and various classifier architectures. The system possesses the ability of dealing with difficult field-collected images of objects and the techniques employed by its component modules make possible its extension to the domain of complex multiple-object 3D scene recognition. The system is expected to find immediate applicability in the field of marine biota classification.

View PDFchevron_right
3D structures for generic object recognition
Alexandre R . J . Francois

Pattern Recognition, 2000. …, 2000

View PDFchevron_right
A learning approach to 3D object representation for classification
Linda Shapiro

2008

In this paper we describe our 3D object signature for 3D object classification. The signature is based on a learning approach that finds salient points on a 3D object and represent these points in a 2D spatial map based on a longitude-latitude transformation. Experimental results show high classification rates on both pose-normalized and rotated objects and include a study on classification accuracy as a function of number of rotations in the training set.

View PDFchevron_right
Appearance-based 3D object recognition with time-invariant features
Elisabetta Delponte

2007

In this paper we explore the interlink between temporally dense view-based object recognition and sparse image representations with local keypoints. The temporal component is an add on that allows us to extract information which is distinctive of a given object in a given view-point range. We use temporal descriptions both for training and for testing. In the training phase each image sequence contains one object only, observed at different view points. At run time video shots are analyzed looking for known objects. Train and test video shots are represented by a structure of scale-space keypoints selected so that they are robust to view-point changes. In the matching phase we emphasize co-occurring keypoints and attenuate the importance of isolated points, both in the model and in the test representation. With our prototype recognition system we obtained very good results in controlled and unconstrained environments, setting the ground for real world applications such as automatic place recognition, or robot object grasping.

View PDFchevron_right

References (55)

  1. S. Agarwal and D. Roth. Learning a sparse representa- tion for object detection. In Proc. European Conf. Comp. Vision, volume LNCS 2353, pages 113-127, Copenhagen, Denmark, 2002.
  2. R. Basri and S. Ullman. The alignment of objects with smooth surfaces. In Proc. Int. Conf. Comp. Vision, pages 482-488, Tampa, FL, 1988.
  3. P.N. Belhumeur, J.P. Hesphanha, and D.J. Kriegman. Eigen- faces vs. Fisherfaces: recognition using class-specific linear projection. In Proc. European Conf. Comp. Vision, pages 45-58, 1996.
  4. S. Belongie, J. Malik, and J. Puzicha. Matching shapes. In Proc. Int. Conf. Comp. Vision, pages 454-461, 2001.
  5. C. Bishop. Neural Networks for Pattern Recognition. Ox- ford University Press, 1995.
  6. T.E. Boult and L.G. Brown. Factorization-based segmenta- tion of motions. In IEEE Workshop on Visual Motion, pages 179-186, 1991.
  7. P. Brodatz. Textures: A Photographic Album for Artists and Designers. Dover, New York, 1966.
  8. R.A. Brooks. Symbolic reasoning among 3-D models and 2- D images. Artificial Intelligence Journal, 17(1-3):285-348, 1981.
  9. J. B. Burns, R. S. Weiss, and E. M. Riseman. View variation of point-set and line-segment features. IEEE Trans. Patt. Anal. Mach. Intell., 15(1):51-68, January 1993.
  10. O. Carmichael and M. Hebert. Object recognition by a cas- cade of edge probes. In British Machine Vision Conf., 2002.
  11. J. Costeira and T. Kanade. A multi-body factorization method for motion analysis. In Proc. Int. Conf. Comp. Vi- sion, pages 1071-1076, Boston, MA, 1995.
  12. R.O. Duda, P.E. Hart, and D.G. Stork. Pattern Classifica- tion. Wiley-Interscience, 2001. Second edition.
  13. O. Faugeras, Q.-T. Luong, and T. Papadopoulo. The Geom- etry of Multiple Images. MIT Press, 2001.
  14. C.W. Gear. Multibody grouping in moving objects. Int. J. of Comp. Vision, 29(2):133-150, August/September 1998.
  15. C. Harris and M. Stephens. A combined edge and corner detector. In 4 th Alvey Vision Conference, pages 189-192, Manchester, UK, 1988.
  16. B. Heisele, T. Serre, M. Pontil, and T. Poggio. Component- based face detection. In Proceedings IEEE Conference Computer Vision and Pattern Recognition, pages 657-662, 2001.
  17. D.P. Huttenlocher and S. Ullman. Object recognition using alignment. In Proc. Int. Conf. Comp. Vision, pages 102-111, London, U.K., June 1987.
  18. A.E. Johnson and M. Hebert. Surface matching for object recognition in complex three-dimensional scenes. Image and Vision Computing, 16:635-651, 1998.
  19. L. Kaufman and P.J. Rousseeuw. Finding Groups in Data. An Introduction to Cluster Analysis. John Wiley and sons, New York, 1990.
  20. J.J. Koenderink and A.J. Van Doorn. The structure of locally orderless images. Int. J. of Comp. Vision, 31(2/3):159-168, 1999.
  21. S. Lazebnik, C. Schmid, and J. Ponce. Affine-invariant local descriptors and neighborhood statistics for texture recogni- tion. In Proc. Int. Conf. Comp. Vision, 2003.
  22. S. Lazebnik, C. Schmid, and J. Ponce. Sparse texture rep- resentations using affine-invariant neighborhoods. In Proc. IEEE Conf. Comp. Vision Patt. Recog., 2003.
  23. T. Lindeberg. Feature detection with automatic scale selec- tion. Int. J. of Comp. Vision, 30(2):79-116, 1998.
  24. T. Lindeberg and J. Gårding. Shape-adapted smoothing in estimation of 3-D shape cues from affine deformations of lo- cal 2-D brightness structure. Image and Vision Computing, 15(6):415-434, 1997.
  25. F. Liu and W. Picard. Periodicity, directionality, and ran- domness: World features for image modeling and retrieval. IEEE Trans. Patt. Anal. Mach. Intell., 18(7):722-733, 1996.
  26. D.G. Lowe. The viewpoint consistency constraint. Int. J. of Comp. Vision, 1(1):57-72, 1987.
  27. S. Mahamud, M. Hebert, and J. Lafferty. Combining simple discriminators for object discrimination. In Proc. European Conf. Comp. Vision, Copenhagen, Denmark, May 2002.
  28. K. Mikolajczyk and C. Schmid. An affine invariant inter- est point detector. In Proc. European Conf. Comp. Vision, volume I, pages 128-142, Copenhagen, Denmark, 2002.
  29. J.L. Mundy and A. Zisserman. Geometric Invariance in Computer Vision. MIT Press, Cambridge, Mass., 1992.
  30. J.L. Mundy, A. Zisserman, and D. Forsyth. Applications of Invariance in Computer Vision, volume 825 of Lecture Notes in Computer Science. Springer-Verlag, 1994.
  31. H. Murase and S. Nayar. Visual learning and recognition of 3D objects from appearance. Int. J. of Comp. Vision, 14(1):5-24, 1995.
  32. V.S. Nalwa. Line-drawing interpretation: bilateral symme- try. In Proc. DARPA Image Understanding Workshop, pages 956-967, Los Angeles, CA, February 1987.
  33. R. Nevatia and T.O. Binford. Description and recognition of complex curved objects. Artificial Intelligence Journal, 8:77-98, 1977.
  34. K Nigam, A. McCallum, S. Thrun, and T. Mitchell. Text classification from labeled and unlabeled documents using EM. Machine Learning, 39(2-3):103-134, 2000.
  35. C.P. Papageorgiou, M. Oren, and T. Poggio. A general framework for object detection. In Proc. Int. Conf. Comp. Vision, pages 555-562, 1998.
  36. R. Picard, T. Kabir, and F. Liu. Real-time recognition with the entire Brodatz texture database. In Proc. IEEE Conf. Comp. Vision Patt. Recog., pages 638-639, New York City, NY, 1993.
  37. C.J. Poelman and T. Kanade. A paraperspective factoriza- tion method for shape and motion recovery. IEEE Trans. Patt. Anal. Mach. Intell., 19(3):206-218, March 1997.
  38. J. Ponce, D. Chelberg, and W. Mann. Invariant properties of straight homogeneous generalized cylinders and their con- tours. IEEE Trans. Patt. Anal. Mach. Intell., 11(9):951-966, September 1989.
  39. R. Ronfard, C. Schmid, and B. Triggs. Learning to parse pictures of people. In Proc. European Conf. Comp. Vision, volume IV, pages 700-714, Copenhagen, Denmark, 2002.
  40. A. Rosenfeld, R. Hummel, and S. Zucker. Scene labeling by relaxation operations. IEEE Trans. on Systems, Man, and Cybernetics, 6(6):420-433, 1976.
  41. F. Rothganger, S. Lazebnik, C. Schmid, and J. Ponce. 3D object modeling and recognition using affine-invariant patches and multi-view spatial constraints. In Proc. IEEE Conf. Comp. Vision Patt. Recog., 2003.
  42. H.A. Rowley, S. Baluja, and T. Kanade. Neural network- based face detection. IEEE Trans. Patt. Anal. Mach. Intell., 20(1):23-38, 1998.
  43. Y. Rubner, C. Tomasi, and L.J. Guibas. A metric for distri- butions with applications to image databases. In Proc. Int. Conf. Comp. Vision, 1998.
  44. F. Schaffalitzky and A. Zisserman. Viewpoint invariant tex- ture matching and wide baseline stereo. In Proc. Int. Conf. Comp. Vision, Vancouver, Canada, 2001.
  45. C. Schmid. Constructing models for content-based image retrieval. In CVPR, 2001.
  46. C. Schmid and R. Mohr. Local grayvalue invariants for image retrieval. IEEE Trans. Patt. Anal. Mach. Intell., 19(5):530-535, May 1997.
  47. H. Schneiderman and T. Kanade. A statistical method for 3D object detection applied to faces and cars. In Proc. IEEE Conf. Comp. Vision Patt. Recog., Hilton Head, SC, 2000.
  48. A. Sethi, D. Renaudie, D.J. Kriegman, and J. Ponce. Curve and surface duals and the recognition of curved 3D objects from their silhouette. Int. J. of Comp. Vision, 2003. In press.
  49. C. Tomasi and T. Kanade. Shape and motion from image streams under orthography: a factorization method. Int. J. of Comp. Vision, 9(2):137-154, 1992.
  50. M. Turk and A.P. Pentland. Face recognition using eigen- faces. J. of Cognitive Neuroscience, 3(1), 1991.
  51. M. Varma and A. Zisserman. Classifying images of mate- rials: Achieving viewpoint and illumination independence. In Proc. European Conf. Comp. Vision, 2002.
  52. P. Viola and M. Jones. Robust real-time object detection. Technical Report CRL 01/01, Compaq Cambridge Research Laboratory, 2001.
  53. M. Weber, M. Welling, and P. Perona. Unsupervised learn- ing of models for recognition. In Proc. European Conf. Comp. Vision, Dublin, Ireland, 2000.
  54. K. Xu, B. Georgescu, D. Comaniciu, and P. Meer. Perfor- mance analysis in content-based retrieval with textures. In Proc. Int. Conf. Patt. Recog., 2000.
  55. M. Zerroug and R. Nevatia. From an intensity image to 3D segmented descriptions. In J. Ponce, A. Zisserman, and M. Hebert, editors, Object Representation in Computer Vi- sion II, number 1144 in Lecture Notes in Computer Sci- ences, pages 11-24. Springer-Verlag, 1996.

Related papers

3D Object Modeling and Recognition from Photographs and Image Sequences
Jean Ponce

Lecture Notes in Computer Science, 2006

This chapter proposes a representation of rigid three-dimensional (3D) objects in terms of local affine-invariant descriptors of their images and the spatial relationships between the corresponding surface patches. Geometric constraints associated with different views of the same patches under affine projection are combined with a normalized representation of their appearance to guide the matching process involved in object modeling and recognition tasks. The proposed approach is applied in two domains: (1) Photographs -models of rigid objects are constructed from small sets of images and recognized in highly cluttered shots taken from arbitrary viewpoints. (2) Video -dynamic scenes containing multiple moving objects are segmented into rigid components, and the resulting 3D models are directly matched to each other, giving a novel approach to video indexing and retrieval.

View PDFchevron_right
Joint utilization of local appearance and geometric invariants for 3D object recognition
Aydin Alatan

Multimedia Tools and Applications, 2013

In this appendix, geometric invariants are derived for point sets in affine spaces. These derivations are performed in a way to enable formation of relations between invariants of point sets in 3D spaces and their projections in 2D. The derivations included in this chapter reproduces the work of Weiss [1]. 1.1 Linear Algebraic Prerequisites Linear algebraic operations constitute an important part of the derivations in Section 1.2. This section is devoted to theorems of linear algebra that are essential for the aforementioned derivations. The first theorem below is related to the nlinearity property of determinants on n × n matrices, and the ones that follow are related to the relations governing the product of determinants and row operations of determinants. 1.1.1 Theorem 1 Determinant is an n-linear function on n × n matrices [2], that is to say, if the i th row of a square matrix is of the form cR i + dR ′ i where R i and R ′ i are 1 × n matrices, then we have:

View PDFchevron_right
Analysis on a Local Approach to 3D Object Recognition
Elisabetta Delponte

2006

We present a method for 3D object modeling and recognition which is robust to scale and illumination changes, and to viewpoint variations. The object model is derived from the local features extracted and tracked on an image sequence of the object. The recognition phase is based on an SVM classifier. We analyse in depth all the crucial steps of the method, and report very promising results on a dataset of 11 objects, that show how the method is also tolerant to occlusions and moderate scene clutter.

View PDFchevron_right
3D Object Recognition Based on Image Features: A Survey
Mohammed M Elmogy

International Journal of Computer and Information Technology (IJCIT), 2014

Object recognition is a basic application domain in computer vision. For many decades, it is considered as an area of extensive research especially in 3D. 3D object recognition can be defined as the task of finding and identifying objects in the real world from an image or a video sequence. It is still a hot research topic in computer vision because it has many challenges such as viewpoint variations, scaling, illumination changes, partial occlusion, and background clutter. Many approaches and algorithms are proposed and implemented to overcome these challenges. In this paper, we will discuss the current computer vision literature on 3D object recognition. We will introduce an overview of the current approaches of some important problems in visual recognition, to analyze their strengths and weaknesses. Finally, we will present particular challenges in 3D object recognition approaches that have been used recently. As well as, possible directions for future research will be presented in this field.

View PDFchevron_right
3D object indexing and recognition
Feryel Souami

Applied Mathematics and …, 2008

In this paper, we address the problem of 3D object recognition from a single 2D image using models database. We propose a method based on geometric quasi-invariant features of the 2D images. We index the 2D images in a model base using a modified quad-tree technique that enhance the research process. The final vote that matches the 2D object image to the 3D object of the database is solved by a vector approximation file which overcomes the difficulties of high dimensionality by following not the data partitioning approach of conventional index methods, but rather as filter based approach.

View PDFchevron_right

Related topics

  • Object Data Model
  • Image Sequence
    • 580 California St., Suite 400
    San Francisco, CA, 94104
    © 2025 Academia. All rights reserved