Reported Results on MPEG-7 Core Experiment CE-Shape-1 Test Set


Improving shape retrieval/matching:

    The retrieval rate is measured by the so-called bull’s eye score. Every shape in the database is compared to all other shapes, and the number of shapes from the same class among the 40 most similar shapes is reported. The bull’s eye retrieval rate is the ratio of the total number of shapes from the same class to the highest possible number (which is 20 × 1400). Thus, the best possible rate is 100%.

TABLE I

Alg.

CSS

 

[40]

Vis. Parts

 

[5]

Shape

Contexts

[2]

Aligning

Curves

[41]

Distance

Set

[42]

Prob.

Approach

[43]

Chance

Prob.

[44]

Skeletal

Context

[45]

Gen.

Model

[3]

Optimized

CSS

[46]

Score
75.44%
76.45%
76.51%
78.16%
78.38%
79.19%
79.36%
79.92%
80.03%
81.12%
Alg.

Contour

Seg.

[47]

Multiscale

Rep.

[48]

Shape

L’AneRouge

[49]

Fixed

Cor.

[50]

Inner

Distance

[4]

Symbolic

Rep.

[30]

Hier.

Procrustes

[8]

Triangle

Area

[29]

Shape

Tree

[9]

Our

Method

[1]

Score
84.33%
84.93%
85.25%
85.40%
85.40%
85.92%
86.35%
87.23%
87.70%
91.61%

Alg.

 

Our

Method

[10]

                 
Score 93.32%                  

 ABOVE ARE THE  RETRIEVAL RATES (BULL’S EYE) OF DIFFERENT METHODS ON THE MPEG-7 DATA SET.

[1] Xiang Bai, Xingwei Yang, Longin Jan Latecki, Wenyu Liu, Zhuowen Tu. Learning Context Sensitive Shape Similarity by Graph Transduction. IEEE Trans. Pattern Analysis and Machine Intelligence (PAMI) 2009.

[2] S. Belongie, J. Malik, and J. Puzicha, “Shape matching and object recognition using shape contexts,” IEEE Trans. PAMI, vol. 24, pp. 705–522, 2002.

[3] Z. Tu and A. L. Yuille, “Shape matching and recognition - using generative models and informative features,” in ECCV, 2004, pp. 195–209.

[4] H. Ling and D. Jacobs, “Shape classification using the inner-distance,” IEEE Trans. PAMI, vol. 29, no. 2, pp. 286–299, 2007.

[5] L. J. Latecki and R. Lak¨amper, “Shape similarity measure based on correspondence of visual parts,” IEEE Trans. PAMI, vol. 22, no. 10, pp. 1185–1190, 2000.

[8] G. McNeill and S. Vijayakumar, “Hierarchical procrustes matching for shape retrieval,” in Proc. CVPR, 2006.

[9] P. F. Felzenszwalb and J. Schwartz, “Hierarchical matching of deformable shapes.” in CVPR, 2007.

[10] Xingwei Yang, Suzan Koknar-Tezel, and Longin Jan Latecki. Locally Constrained Diffusion Process on Locally Densified Distance Spaces with Applications to Shape Retrieval. CVPR 2009.

[29] N. Alajlan, M. Kamel, and G. Freeman, “Geometry-based image retrieval in binary image databases,” IEEE Trans. on PAMI, vol. 30, no. 6, pp. 1003–1013, 2008.

[30] M. Daliri and V. Torre, “Robust symbolic representation for shape recognition and retrieval,” Pattern Recognition, vol. 41, no. 5, pp. 1799–1815, 2008.

[40] F. Mokhtarian, F. Abbasi, and J. Kittler, “Efficient and robust retrieval by shape content through curvature scale space,” Image Databases and Multi-Media Search, A.W.M Smeulders and R. Jain eds, pp. 51–58, 1997.

[41] T. Sebastian, P. Klein, and B. Kimia, “On aligning curves,” IEEE Trans. PAMI, vol. 25, pp. 116–125, 2003.

[42] C. Grigorescu and N. Petkov, “Distance sets for shape filters and shape recognition,” IEEE Trans. on Image Processing, vol. 12, no. 7, pp. 729–739, 2003.

[43] G. McNeill and S. Vijayakumar, “2d shape classification and retrieval,” in IJCAI, 2005.

[44] B. Super, “Learning chance probability functions for shape retrieval or classification,” in Proceedings of the IEEE Workshop on Learning in CVPR, 2004.

[45] J. Xie, P. Heng, and M. Shah, “Shape matching and modeling using skeletal context,” Pattern Recognition, vol. 41, no. 5, pp. 1756–1767, 2008.

[46] F. Mokhtarian and M. Bober, Curvature Scale Space Representation: Theory, Applications & MPEG-7 Standardization. Dordrecht: Kluwer Academic Publishers, 2003.

[47] E. Attalla and P. Siy, “Robust shape similarity retrieval based on contour segmentation polygonal multiresolution and elastic matching,” Pattern Recognition, vol. 38, no. 12, pp. 2229–2241, 2005.

[48] T. Adamek and N. O’Connor, “A multiscale representation method for nonrigid shapes with a single closed contour,” IEEE Trans. on CSVT, vol. 14, no. 5, pp. 742–753, 2004.

[49] A. Peter, A. Rangarajan, and J. Ho, “Shape l’ˆane rouge: Sliding wavelets for indexing and retrieval,” in CVPR, 2008.

[50] B. Super, “Retrieval from shape databases using chance probability functions and fixed correspondence,” Int. J. Pattern Recognition Artif. Intell., vol. 20, no. 8, pp. 1117–1137, 2006.

Designed by: Richard Ralph