Perceptual Grouping Based on Iterative Multi-scale Tensor Voting
We propose a new approach for perceptual grouping of oriented segments in highly cluttered images based on tensor voting. Segments are represented as second-order tensors and communicate with each other through a voting scheme that incorporates the Gestalt principles of visual perception. An iterative scheme has been devised which removes noise segments in a conservative way using multi-scale analysis and re-voting. We have tested our approach on data sets composed of real objects in real backgrounds. Our experimental results indicate that our method can segment successfully objects in images with up to twenty times more noise segments than object ones.
KeywordsReceiver Operational Characteristic Curve Perceptual Grouping Input Token Ambiguity Region Benchmark Image
Unable to display preview. Download preview PDF.
- 2.Ahuja, N., Tuceryan, M.: Extraction of early perceptual structure in dot patterns: integrating region, boundary, and component gestalt. In: CVGIP, pp. 304–356 (1989)Google Scholar
- 3.Mohan, R., Nevatia, R.: Segmentation and description based on perceptual organization. In: CVPR, pp. 333–341 (1989)Google Scholar
- 4.Ullman, S.: Filling-in the gaps: The shape of subjective contours and a model for their generation. Biological Cybernetics 25 (1976)Google Scholar
- 5.Parent, P., Zucker, S.W.: Trace inference, curvature consistency, and curve detection. IEEE Trans. PAMI 11, 823–839 (1989)Google Scholar
- 6.Ullman, S., Sha’ashua, A.: Structural saliency: The detection of globally salient structures using a locally connect network. In: 2nd. International Conference on Computer Vision - ICCV 1988 (1988)Google Scholar
- 10.Medioni, G., Kang, S.B.: A Computational Framework for Segmentation and Grouping. In: The Tensor Voting Framework. Prentice Hall, Englewood Cliffs (2005)Google Scholar