主要研究计算机视觉中一个非常具有挑战性的问题——自然图像中对称轴的检测. 由于图像中杂乱的场景和物体形态的变化, 使得在自然图像中判断像素是否处于对称轴上是非常困难的. 为了解决这个难题, 考虑到边缘与对称轴的互补关系, 提出了2种边缘特征用于帮助对称轴的检测. 2种特征都定义在成对的边缘上, 分别是为了找到边缘强度高和到对称轴距离相等的成对的边缘. 在多尺度和多角度上提取这2种边缘特征, 把它们与底层描述子(颜色、亮度、纹理等)差分特征结合在一起, 在多示例学习的框架下检测自然图像中的对称轴.在SYMMAX300数据集上的实验结果证明了2种边缘特征能够提升对称轴检测的性能.
This paper studies symmetry detection in natural images, which is a challenging problem in computer vision. To differentiate symmetry and non-symmetry in natural images is intractable due to the large variation in objects and the cluttered scene. To address this problem, two types of edge features motivated by the fact that symmetries are complementary to edges are proposed. These two types of features are both defined on pairs of edges to search pairs of edges with consistent high strength and equal distances to symmetries, respectively. The proposed edge features at multiple scales and orientations
and integrate them with low level cues (color, brightness and texture) under a multiple instance learning framework to detect symmetries are extracted. The experimental results on SYMMAX300 dataset demonstrate that both proposed edge features can improve performance of symmetry detection.
[1] Bai X, Latecki L J, Liu W. Skeleton pruning by contour partitioning with discrete curve evolution [J]. IEEE Trans PAMI, 2007, 29(3): 449-462.
[2] Bai X, Latecki L J. Path similarity skeleton graph matching [J]. IEEE Trans PAMI, 2008,30(5): 1282-1292.
[3] Bai X, Wang X, Latecki L J, et al. Active skeleton for non-rigid object detection [C]// International Conference on Computer Vision. 2009: 575-582.
[4] Siddiqi K, Pizer S. Medial representations [M]. Berlin: Springer-Verlag, 2009.
[5] Tsogkas S, Kokkinos I. Learning-based symmetry detection in natural images [C]// European Conference on Computer Vision. 2012: 41-45.
[6] Martin D, Fowlkes C, Tal D, et al. A database of human segmented natural images and its application to evaluating segmentation algorithms and measuring ecological statistics [C]//
International Conference on Computer Vision. 2001: 416-423.
[7] Martin D, Fowlkes C, Malik J. Learning to detect natural image boundaries using local brightness, color, and texture cues [J]. IEEE Trans PAMI, 2004, 26(5): 530-549.
[8] 刘超, 安平, 赵冰, 等. 基于边缘的面向虚拟视绘制的深度编码算法[J]. 上海大学学报: 自然科学版,2014, 20(4): 450-457.
[9] Rubner Y, Puzicha J, Tomasi C, et al. Empirical evaluation of dissimilarity measures for color and texture [C]// International Conference on Computer Vision. 1999: 1165-1172.
[10] Ren X. Multi-scale improves boundary detection in natural images [C]// European Conference on Computer Vision. 2008: 533–545.
[11] Ren X, Bo L. Discriminatively trained sparse code gradients for contour detection [C]// Annual Conference on Neural Information Processing Systems. 2012: 593-601.
[12] Lindeberg T. Edge detection and ridge detection with automatic scale selection [J]. International Journal of Computer Vision, 1998, 30(2): 117-156.
[13] Levinshtein A, Dickinson S, Sminchisescu C. Multiscale symmetric part detection and grouping [C]// International Conference on Computer Vision. 2009: 2162-2169.
