Bilateral Symmetry Detection and Segmentation via Symmetry-Growing

We present a novel and robust method for localizing and segmenting bilaterally symmetric patterns from real-world images. On the basis of symmetrically matched pairs of local features, the method expands and merges confident local symmetric region matches by exploiting both photometric similarity and geometric consistency via our symmetry-growing framework. It overcomes the limitations of the previous local-feature based approaches by efficiently exploring the image space to grow symmetry beyond the detected symmetric features. The experimental evaluation demonstrates that our method successfully detects the entire regions of multiple symmetric patterns from real-world images, and clearly outperforms the state-of-the-art methods in accuracy and robustness.

• Type: Conference
• Year: 2009
• Where: BMVC
• Authors: Minsu Cho (Seoul National University), Kyoung Mu Lee (Seoul National University)
• Keywords: Bilateral Symmetry Detection, BMVC 2009, Segmentation, Symmetry-Growing