13.3.13 Structural Matching for Computer Vision

Chapter Contents (Back)
Matching, Structures. A lot of overlap with:
See also General Structure and Graph Representation, Relations, Neighbors.

Barrow, H.G., and Popplestone, R.J.,
Relational Descriptions in Picture Processing,
MI(VI), 1971, pp. 377-396. Matching, Tree Search. Classical work in structural description, matching and segmentation. The region growing technique is intended to be an incomplete, fast region grower. The basic idea is to collect points that are similar (within 3 gray levels out of a total of 16) to preselected grid points (a 16X16 grid over the original 64X64 image). These elementary regions may overlap. These elementary regions are merged according to the contrast along the border. This procedure also discards background regions (i.e. those which touch the sides of the image). The simple region grower produces the basic descritpion of the object. A structural (graph-based) description is generated from properties of the regions (brightness and shape) and relations between regions (adjacency, bigger, distance between, and positional relations). The correspondence between the model graph and the resulting image graph is determined by a branch-and-bound tree searching technique. See related segmentation work:
See also Scene Analysis Using Regions. BibRef 7100

Barrow, H.G., Ambler, A.P., and Burstall, R.M.,
Some Techniques for Recognizing Structures in Pictures,
FPR72(1-29). BibRef 7200 CMetImAly77(397-425). Matching, Graphs. Recognize Structures. Another early classical work in structural matching. BibRef

Ambler, A.P., Popplestone, R.J.,
Inferring the Position of Bodies from Specified Spatial Relationships,
AI(6), No. 2, June 1975, pp. 157-174.
Elsevier DOI BibRef 7506

Popplestone, R.J., Ambler, A.P., and Bellos, I.M.,
An Interpreter for a Language for Describing Assemblies,
AI(14), No. 1, August 1980, pp. 79-107.
Elsevier DOI BibRef 8008

Barrow, H.G., and Burstall, R.M.,
Subgraph Isomorphism, Matching Relational Structures and Maximal Cliques,
IPL(4), 1976, pp. 83-84. Association Graph. BibRef 7600

Harlow, C.A.[Charles A.],
Image Analysis and Graphs,
CGIP(2), No. 1, August 1973, pp. 60-82.
Elsevier DOI BibRef 7308

Cheng, J.K., and Huang, T.S.,
Image Registration by Matching Relational Structures,
PR(17), No. 1, 1984, pp. 149-159.
Elsevier DOI BibRef 8400
Earlier: ICPR82(354-356). BibRef
Earlier: PRIP81(542-547). Recognize Structures. Recognition is based on long edge segments of tools. Binary relations are combined to produce ternary relations. Construct relational descriptions of the boundary - very gross features - long edges and chords of curved sections. The matching is based on "stars" - nodes and every thing it is linked to, the compatibility is overlap of possible assignments in pairs of stars. There is a star for each node, with ratings for others stars, refinement through relaxation (10 to 20 iterations). BibRef

Cheng, J.K., and Huang, T.S.,
A Subgraph Isomorphism Algorithm Using Resolution,
PR(13), No. 5, 1981, pp. 371-379.
Elsevier DOI BibRef 8100

Cheng, J.K., and Huang, T.S.,
Recognition of Curvilinear Objects by Matching Relational Structures,
PRIP82(343-348). BibRef 8200

Bolles, R.C., and Cain, R.A.,
Recognizing and Locating Partially Visible Objects: The Local-Feature-Focus Method,
IJRR(1), No. 3, Fall 1982, pp. 57-82. BibRef 8200
Earlier: A1 only: AAAI-80(41-43). BibRef
And: A1, A2:
Recognizing and Locating Partially Visible Workpieces: The Local-Feature-Focus Method,
PRIP82(498-503). Recognize Two-Dimensional Objects. The location and matching is based on distinct features (holes and corners) with a hypothesis formed from several consistent feature locations. Testing the hypothesis yields more features and an indication of occlusions. Find a feature, predict the location of other features, (cluster) graph matching in applied to identify the cluster, and then extract the rest of the object to verify the identification. BibRef

Bolles, R.C.,
Verification Vision for Programmable Assembly,
IJCAI77(569-575). BibRef 7700
And:
Verification Vision within a Programmable Assembly System,
Stanford AI275, December 1975. BibRef

Kupeev, K.Y., Wolfson, H.J.,
A New Method of Estimating Shape Similarity,
PRL(17), No. 8, July 1 1996, pp. 873-887. 9608
Perceptual similarity using contours.
PS File. BibRef

Kupeev, K.Y., and Wolfson, H.J.,
On Shape Similarity,
ICPR94(A:227-231).
IEEE DOI Perceptual similarity of contours for shape matching. BibRef 9400

Fitch, A.J., Kadyrov, A., Christmas, W.J., Kittler, J.V.,
Fast robust correlation,
IP(14), No. 8, August 2005, pp. 1063-1073.
IEEE DOI 0508
BibRef
Earlier:
Fast exhaustive robust matching,
ICPR02(III: 903-906).
IEEE DOI 0211
BibRef

Ommer, B.[Björn], Mader, T.[Theodor], Buhmann, J.M.[Joachim M.],
Seeing the Objects Behind the Dots: Recognition in Videos from a Moving Camera,
IJCV(83), No. 1, June 2009, pp. xx-yy.
Springer DOI 0903
BibRef

Ommer, B.[Bjorn], Buhmann, J.M.[Joachim M.],
Learning the Compositional Nature of Visual Object Categories for Recognition,
PAMI(32), No. 3, March 2010, pp. 501-516.
IEEE DOI 1002
BibRef
Earlier:
Learning the Compositional Nature of Visual Objects,
CVPR07(1-8).
IEEE DOI 0706
BibRef
And:
Compositional Object Recognition, Segmentation, and Tracking in Video,
EMMCVPR07(318-333).
Springer DOI 0708
BibRef
Earlier:
Learning Compositional Categorization Models,
ECCV06(III: 316-329).
Springer DOI 0608
BibRef
Earlier:
Object Categorization by Compositional Graphical Models,
EMMCVPR05(235-250).
Springer DOI 0601
Learn composition of objects (of parts) BibRef

Ommer, B.[Bjorn], Sauter, M.[Michael], Buhmann, J.M.[Joachim M.],
Learning Top-Down Grouping of Compositional Hierarchies for Recognition,
PercOrg06(194).
IEEE DOI 0609
BibRef

Roth, V.[Volker], Ommer, B.[Björn],
Exploiting Low-Level Image Segmentation for Object Recognition,
DAGM06(11-20).
Springer DOI 0610
BibRef


Zhou, M.[Mohan], Bai, Y.L.[Ya-Long], Zhang, W.[Wei], Zhao, T.J.[Tie-Jun], Mei, T.[Tao],
Look-Into-Object: Self-Supervised Structure Modeling for Object Recognition,
CVPR20(11771-11780)
IEEE DOI 2008
Object recognition, Visualization, Training, Task analysis, Image recognition, Computational modeling, Neural networks BibRef

Sadeghi, F.[Fereshteh], Tappen, M.F.[Marshall F.],
Latent Pyramidal Regions for Recognizing Scenes,
ECCV12(V: 228-241).
Springer DOI 1210
BibRef

Silva, F.B.[Fernanda B.], Tabbone, S.[Salvatore], da Silva Torres, R.[Ricardo],
BoG: A New Approach for Graph Matching,
ICPR14(82-87)
IEEE DOI 1412
Accuracy; Dictionaries; Kernel; Training; Vectors; Visualization; Vocabulary BibRef

Penatti, O.A.B.[Otávio A. B.], Valle, E.[Eduardo], da Silva Torres, R.[Ricardo],
Encoding Spatial Arrangement of Visual Words,
CIARP11(240-247).
Springer DOI 1111
BibRef

Peralta, B.[Billy], Soto, A.[Alvaro],
Mixing Hierarchical Contexts for Object Recognition,
CIARP11(232-239).
Springer DOI 1111
Category level recognition. BibRef

Yao, B.P.[Bang-Peng], Niebles, J.C.[Juan Carlos], Fei-Fei, L.[Li],
Mining discriminative adjectives and prepositions for natural scene recognition,
VCL-ViSU09(100-106).
IEEE DOI 0906
Appearance and relations of patches. BibRef

Shokoufandeh, A.[Ali], Dickinson, S.J.[Sven J.], Jönsson, C.[Clas], Bretzner, L.[Lars], Lindeberg, T.[Tony],
On the Representation and Matching of Qualitative Shape at Multiple Scales,
ECCV02(III: 759 ff.).
Springer DOI 0205
Initial match of low level features. BibRef

Yamaguchi, A.[Akashi], Inokuchi, S.[Seiji], Kochi, K.[Kazutaka],
Stereo Matching for Stone Statues Using SRI Parameters and Relational Graph,
ICPR98(Vol I: 785-787).
IEEE DOI 9808
BibRef

Dubuisson-Jolly, M.P.[Marie-Pierre], Jain, A.K.,
A Modified Hausdorff Distance for Object Matching,
ICPR94(A:566-568).
IEEE DOI BibRef 9400

Enomoto, H., Yonezaki, N., Nitta, K.,
A Model for Perception of Structural Image Feature,
IJCAI79(257-259). BibRef 7900

Chapter on Matching and Recognition Using Volumes, High Level Vision Techniques, Invariants continues in
Object Recognition, General Techniques .


Last update:Mar 16, 2024 at 20:36:19