Skip to main content
SpringerLink
Log in

Object detection based on scale-invariant partial shape matching

  • Original Paper
  • Published:
Machine Vision and Applications Aims and scope Submit manuscript

Abstract

This paper aims at detecting objects via a partial shape matching in unlabeled real images. As both the scale and consistent fragment extraction are troublesome issues in computer vision, we first extract the corresponding parts of pairs of matching fragments generated by the curvature extreme points in object contours. Then, we establish the scale-calculable shape descriptor to keep that the partial shape matching algorithm is scale and rotation invariant. In detection stage, a weighted voting scheme is used to locate candidate object centers and followed by a refinement process to obtain the precise object boundaries. Experiments on ETHZ shape category database validate that using single model shape without training for each category can match (or exceed) the performance of state-of-the-art object detection algorithms.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6
Fig. 7
Fig. 8
Fig. 9

Similar content being viewed by others

References

  1. Soysal, Ö., Chen, J.: Object recognition by spectral feature derived from canonical shape representation. Mach. Vis. Appl. 24(2), 855–868 (2013)

  2. Kouzani, A.: Classification of face images using local iterated function systems. Mach. Vis. Appl. 19(4), 223–248 (2008)

    Article  Google Scholar 

  3. Demirci, M.: Retrievin 2d shapes using caterpillar decomposition. Mach. Vis. Appl. 24(2), 435–445 (2013)

  4. Cao, Y., Zhang, Z., Czogiel, I., Dryden, I., Wang, S.: 2d nonrigid partial shape matching using mcmc and contour subdivision. In: IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp. 2345–2352 (2011)

  5. Smith, R., Pham, B.: A robust object category detection system using deformable shapes. Mach. Vis. Appl. 20(2), 119–130 (2009)

    Article  Google Scholar 

  6. Ferrari, V., Jurie, F., Schmid, C.: From images to shape models for object detection. IJCV 87(3), 284–303 (2010)

    Article  Google Scholar 

  7. Riemenschneider, H., Donoser, M., Bischof, H.: Using partial edge contour matches for efficient object category localization. In: Computer Vision–ECCV, pp. 29–42 (2010)

  8. Sun, K., Super, B.: Classification of contour shapes using class segment sets. CVPR 2, 727–733 (2005)

    Google Scholar 

  9. Latecki, L., Megalooikonomou, V., Wang, Q., Yu, D.: An elastic partial shape matching technique. Pattern Recognit. 40(11), 3069–3080 (2007)

    Article  MATH  Google Scholar 

  10. Ferrari, V., Fevrier, L., Jurie, F., Schmid, C.: Groups of adjacent contour segments for object detection. In: IEEE Transactions on Pattern Analysis and Machine Intelligence, pp. 36–51 (2007)

  11. Felzenszwalb, P., McAllester, D., Ramanan, D.: A discriminatively trained, multiscale, deformable part model. In: IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp. 1–8 (2008)

  12. Srinivasan, P., Zhu, Q., Shi, J.: Many-to-one contour matching for describing and discriminating object shape. In: IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp. 1673–1680 (2010)

  13. Jurie, F., Schmid, C.: Scale-invariant shape features for recognition of object categories. In: IEEE Conference on Computer Vision and Pattern Recognition (CVPR), vol. 2, pp. 90–96 (2004)

  14. Ma, T., Latecki, L.: From partial shape matching through local deformation to robust global shape similarity for object detection. In: IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp. 1441–1448 (2011)

  15. Fergus, R., Perona, P., Zisserman, A.: Object class recognition by unsupervised scale-invariant learning. In: IEEE Conference on Computer Vision and Pattern Recognition (CVPR), vol. 2, pp. 264–271 (2003)

  16. Ferrari, V., Jurie, F., Schmid, C.: Accurate object detection with deformable shape models learnt from images. In: IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp. 1–8 (2007)

  17. Maji, S., Berg, A.: Max-margin additive classifiers for detection. In: IEEE 12th International Conference on Computer Vision, pp. 40–47 (2009)

  18. Blaschko, M., Lampert, C.: Learning to localize objects with structured output regression. In: Computer Vision–ECCV, pp. 2–15 (2008)

  19. Bai, X., Li, Q., Latecki, L., Liu, W., Tu, Z.: Shape band: a deformable object detection approach. In: IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp. 1335–1342 (2009)

  20. Jiang, T., Jurie, F., Schmid, C.: Learning shape prior models for object matching. In: IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp. 848–855 (2009)

  21. Belongie, S., Malik, J., Puzicha, J.: Shape matching and object recognition using shape contexts. In: IEEE Transactions on Pattern Analysis and Machine Intelligence, pp. 509–522 (2002)

  22. Ravishankar, S., Jain, A., Mittal, A.: Multi-stage contour based detection of deformable objects. In: Computer Vision–ECCV, pp. 483–496 (2008)

  23. Zhu, Q., Wang, L., Wu, Y., Shi, J.: Contour context selection for object detection: A set-to-set contour matching approach. In: Computer Vision–ECCV, pp. 774–787 (2008)

  24. Lu, C., Latecki, L., Adluru, N., Yang, X., Ling, H.: Shape guided contour grouping with particle filters. In: IEEE 12th International Conference on Computer Vision, pp. 2288–2295 (2009)

  25. Wang, X., Bai, X., Ma, T., Liu, W., Latecki, L.: Fan shape model for object detection. In: IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp. 151–158 (2012)

  26. Martin, D., Fowlkes, C., Malik, J.: Learning to detect natural image boundaries using local brightness, color, and texture cues. In: IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 26, 530–549 (2004)

  27. Hoffman, D., Singh, M.: Salience of visual parts. Cognition 63(1), 29–78 (1997)

    Article  MATH  Google Scholar 

  28. He, X., Yung, N.: Corner detector based on global and local curvature properties. Opt. Eng. 47, 057008 (2008)

    Article  Google Scholar 

  29. Yao, A., Gall, J., Van Gool, L.: A hough transform-based voting framework for action recognition. In: IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp. 2061–2068 (2010)

  30. Neubeck, A., Van Gool, L.: Efficient non-maximum suppression. In: 18th International Conference on Pattern Recognition, vol. 3, pp. 850–855 (2006)

  31. Maji, S., Malik, J.: Object detection using a max-margin hough transform. In: IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp. 1038–1045 (2009)

Download references

Acknowledgments

This work has been funded by Natural Science Foundation of China under Grant Nos. 61401455, 61333019 and 61375014.

Author information

Authors and Affiliations

  1. State Key Laboratory of Robotics, Shenyang Institute of Automation, Chinese Academy of Science, Shenyang, 110016, China

    Huijie Fan, Yang Cong & Yandong Tang

Authors
  1. Huijie Fan
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Yang Cong
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Yandong Tang
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Huijie Fan.

Appendix

Appendix

Here we prove Theorem 1. Figure 3 shows two arbitrary straight lines start from the origin O and intersect two aligned fragments \(\mathbf s ^\prime _i\) and \(\mathbf t ^\prime _j\) at points \(A_1\), \(A_j\), \(B_1\) and \(B_j\). \(C_1\) and \(C_2\) are object centers. We prove that if \(\mathbf s ^\prime _i\) and \(\mathbf t ^\prime _j\) are similar and the scale is s, then

$$\begin{aligned} \frac{|OB_1|}{|OA_1|}= \cdots =\frac{|OB_j|}{|OA_j|}= \cdots =s. \end{aligned}$$

We first prove \(C_2\), \(C_1\) and O are co-linearity. Suppose \(\tilde{C_1}\) instead of \(C_1\) is the object center, since the alignment, which is a congruent transformation, will not alter the relationships of corresponding points on object, i.e., \(\angle XO\tilde{C_1} = \angle XOC_2\), \(\Rightarrow C_2\), \(\tilde{C_1}\) and O are co-linearity.

\(\angle XOA_1 = \angle XOB_1, \Rightarrow A_1\) and \(B_1\) are corresponding points.

$$\begin{aligned}&\Rightarrow \Delta A_1C_1O \backsim \Delta B_1C_2O,\quad \Rightarrow \frac{|OB_1|}{|OA_1|}=\frac{|OC_2|}{|OC_1|} = s.\\&\Rightarrow \frac{|OB_1|}{|OA_1|}= \cdots =\frac{|OB_j|}{|OA_j|}= \cdots =\frac{|OC_2|}{|OC_1|}=s. \end{aligned}$$

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Fan, H., Cong, Y. & Tang, Y. Object detection based on scale-invariant partial shape matching. Machine Vision and Applications 26, 711–721 (2015). https://doi.org/10.1007/s00138-015-0693-y

Download citation

  • Received:

  • Revised:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s00138-015-0693-y

Keywords

Search

Navigation