An Improved Representation of Junctions Through Asymmetric Tensor Diffusion

  • Shawn Arseneau
  • Jeremy R. Cooperstock
Part of the Lecture Notes in Computer Science book series (LNCS, volume 4291)


Junctions form critical features in motion segmentation, image enhancement, and object classification to name but a few application domains. Traditional approaches to identifying junctions include convolutional methods, which involve considerable tuning to handle non-trivial inputs and diffusion techniques that address only symmetric structure. A new approach is proposed that requires minimal tuning and can distinguish between the basic, but critically different, ‘X’ and ‘T’ junctions. This involves a multi-directional representation of gradient structure and employs asymmetric tensor diffusion to emphasize such junctions. The approach combines the desirable properties of asymmetry from convolutional methods with the robustness of local support from diffusion.


Structure Tensor Gradient Structure Motion Segmentation Improve Representation Tensor Vote 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Nicolescu, M., Medioni, G.: Motion segmentation with accurate boundaries - a tensor voting approach. In: Proc. IEEE Conference on Computer Vision and Pattern Recognition, vol. 1, pp. 382–389 (2003)Google Scholar
  2. 2.
    Tschumperlé, D., Deriche, R.: Diffusion pde’s on vector-valued images. IEEE Signal Processing Magazine, 16–25 (2002)Google Scholar
  3. 3.
    Bolles, R., Baker, H., Marimont, D.: Epipolar-plane image analysis: An approach to determining structure from motion. International Journal of Computer Vision 1, 7–55 (1987)CrossRefGoogle Scholar
  4. 4.
    Jiang, X.: On orientation and anisotropy estimation for online fingerprint authentication. IEEE Trans. On Signal Processing 53, 4038–4049 (2005)CrossRefGoogle Scholar
  5. 5.
    Rao, A., Jain, R.: Computerized flow field analysis: Oriented texture fields. IEEE Trans. On Pattern Analysis and Machine Intelligence 14, 693–709 (1992)CrossRefGoogle Scholar
  6. 6.
    Jähne, B.: Spatio-Temporal Image Processing: Theory and Scientific Applications, vol. 751. Springer, Berlin (1993)MATHGoogle Scholar
  7. 7.
    Adelson, E., Bergen, J.: Spatiotemporal energy models for the perception of motion. Journal of the Optical Society of America 2, 284–299 (1985)CrossRefGoogle Scholar
  8. 8.
    Freeman, W., Tanaka, K., Ohta, J., Kyuma, K.: Computer vision for computer games. In: 2nd International Conference on Automatic Face and Gesture Recognition, pp. 100–105 (1996)Google Scholar
  9. 9.
    Iverson, L., Zucker, S.: Logical/linear operators for image curves. IEEE Trans. on Pattern Analysis and Machine Intelligence 17, 982–996 (1995)CrossRefGoogle Scholar
  10. 10.
    Perona, P.: Steerable-scalable kernels for edge detection and junction analysis. In: Proc. European Conference on Computer Vision, pp. 3–18 (1992)Google Scholar
  11. 11.
    Simoncelli, E., Farid, H.: Steerable wedge filters for local orientation analysis. IEEE Trans. On Image Processing 5, 1377–1382 (1996)CrossRefGoogle Scholar
  12. 12.
    Malik, J., Belongie, S., Shi, J., Leung, T.: Textons, contours and regions: Cue integration in image segmentation. In: Proc. IEEE International Conference on Computer Vision, vol. 2, pp. 918–925 (1999)Google Scholar
  13. 13.
    Heeger, D.: Optical flow using spatiotemporal filters. International Journal of Computer Vision 1, 279–302 (1988)CrossRefGoogle Scholar
  14. 14.
    Yu, W., Daniilidis, K., Sommer, G.: Rotated wedge averaging method for junction characterization. In: Proc. IEEE Computer Vision and Pattern Recognition, pp. 390–395 (1998)Google Scholar
  15. 15.
    Michelet, F., Germain, C., Baylou, P., Costa, J.D.: Local multiple orientation estimation: Isotropic and recursive oriented network. In: Proc. IEEE International Conference on Pattern Recognition, vol. 1, pp. 712–715 (2004)Google Scholar
  16. 16.
    Perona, P., Malik, J.: Scale-space and edge detection using anisotropic diffusion. IEEE Trans. Pattern Anal. Machine Intell. 33, 629–639 (1990)CrossRefGoogle Scholar
  17. 17.
    Perona, P.: Orientation diffusions. IEEE Trans. On Image Processing 7, 457–467 (1998)CrossRefGoogle Scholar
  18. 18.
    Tang, B., Sapiro, G., Caselles, V.: Direction diffusion. In: International Conference on Computer Vision, pp. 1245–1252 (1999)Google Scholar
  19. 19.
    Lowe, D.: Distinctive image features from scale-invariant keypoints. International Journal of Computer Vision 60, 91–110 (2004)CrossRefGoogle Scholar
  20. 20.
    Tong, W., Tang, C., Mordohai, P., Medioni, G.: First order augmentation to tensor voting for boundary inference and multiscale analysis in 3d. IEEE Trans. On Pattern Analysis and Machine Intelligence 26, 594–611 (2004)CrossRefGoogle Scholar
  21. 21.
    Parent, P., Zucker, S.: Trace inference, curvature consistency, and curve detection. IEEE Trans. On Pattern Analysis and Machine Intelligence 11, 823–839 (1989)CrossRefGoogle Scholar
  22. 22.
    Ben-Shahar, O., Zucker, S.: Hue fields and color curvatures: A perceptual organization approach to color image denoising. IEEE Computer Vision and Pattern Recognition 2, 713–720 (2003)Google Scholar
  23. 23.
    Savadjiev, P., Campbell, J., Pike, G., Siddiqi, K.: 3d curve inference for diffusion mri regularization. In: International Conference on Medical Image Computing and Computer Assisted Intervention, pp. 123–130 (2005)Google Scholar
  24. 24.
    anonymous for peer review: An asymmetrical diffusion framework for junction analysis. In: British Machine Vision Conference (to appear, 2006)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2006

Authors and Affiliations

  • Shawn Arseneau
    • 1
  • Jeremy R. Cooperstock
    • 1
  1. 1.Centre for Intelligent MachinesMcGill UniversityMontrealCanada

Personalised recommendations