直推式多视图协同分割

朱云峰; 章毓晋

doi:10.3724/SP.J.1146.2010.00839

直推式多视图协同分割

doi: 10.3724/SP.J.1146.2010.00839

基金项目:

国家自然科学基金(60872084)资助课题

计量
- 文章访问数: 3870
- HTML全文浏览量: 98
- PDF下载量: 1483
- 被引次数: 0
出版历程
- 收稿日期: 2010-08-09
- 修回日期: 2011-01-14
- 刊出日期: 2011-04-19

Transductive Co-segmentation of Multi-view Images

摘要

摘要: 多视角拍摄条件下获取到含有同一刚性或静态目标多幅图像后，快速准确地在所有图像中分割出该目标是一个尚未被关注的问题。该文首次给出这个问题的数学描述，并借助图模型的描述方法，推导了它与传统的基于图割分割问题之间的关系。在求解该问题时，该文提出了迭代式协同直推优化算法，算法思想是将优化目标函数拆分为图像空间和3维空间分割两个子优化问题，利用图割和谱聚类分割的方法迭代求解两个子优化问题，并在迭代过程中设计传播、过滤、投票处理引入两个空间标签一致性约束，算法收敛时同时得到图像空间和3维空间的分割结果。最后，使用实际图像序列实验得出算法的平均误判率为3.4%，比较讨论和改进思路一并给出。
- 图像处理 /
- 直推式学习 /
- 图割 /
- 谱聚类分割 /
- 多视角协同分割
Abstract: Fast and efficient segmentation of rigid or stable object in multi-view images is still a unsolved problem. In this paper, the problem is formulated and the relationships between it and traditional min-cut based segmentation problems are also deduced with Graph representation. To minimize the energy function, a novel algorithm named Interactive Transductive Co-Segmentation (ITC-Seg) is proposed. In ITC-Seg, the function is divided into two sub problems which are solved with graph cuts and spectral segmentation methods. Moreover, propagation, filter, voting methods are introduced into the iteration between multi-view images segmentation and qusi-sparse 3D points segmentation, they are used to combine the sub problems with a global label consistent constrains. Finally, the experiments in several images show the error rate of ITC-Seg is 3.4%, discussions and future improvements of the method are also given.
- Image processing /
- Transductive learning /
- Graph cuts /
- Spectral segmentation /
- Multi-view co-segmentation

HTML全文

参考文献(1)

Batra D, Kowdle A, Parikh D, Luo Jiebo, and Chen Tsuhan. iCoseg: interactive co-segmentation with intelligent scribble guidance[C].Computer Vision and Pattern Recognition (CVPR '10), San Francisco, USA, 13-18 June., 2010: 1-8.[2] Agarwal S, Snavely N, Simon I, Seitz S M, and Szeliski R. Building rome in a day[C]. 12th IEEE Intl Conf. Computer Vision(ICCV09), Kyoto, Japan, Sept 29-Oct 2, 2009: 72-79.[3] Furukawa Y and Ponce J. Carved visual hulls for image-based modeling[J]. International Journal of Computer Vision, 2009, 81(1): 53-67.[4] Liu Ye-bin, Dai Qiong-hai, and Xu Wen-li. Continuous depth estimation for multi-view stereo[C]. IEEE Intl Conf. Computer Vision and Pattern Recognition (CVPR09), Florida, USA, June 2009: 2121-2128.[5] Boykov Y and Jolly M P. Interactive graph cuts for optimal boundary and region segmentation of objects in n-d images[C]. ICCV01, Vancouver, British Columbia, Canada, 7-14 July, 2001: 105-112.[6] Grady L. Random walks for image segmentation[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2006, 28(11): 1768-1783.[7] Rother C, Kolmogorov V, and Blake A. Grabcut: interactive foreground extraction using iterated graph cuts[C]. International conference on computer graphics and interactive techniques(SIGGRAPH04), Los Angeles, USA, 10-12 August, 2004: 309-314.[8] Duchenne O, Audibert J Y, Keriven R, Ponce J, and Segonne F. Segmentation by transduction[C]. CVPR08, Anchorage, Alaska, USA, 23-28 June, 2008: 1-8.[9] Meila M and Shi Jian-bo. A random walks view of spectral segmentation[C]. International Conference on AI and Statistics, Key West, Florida, USA, 4-7 January, 2001: 1-8.[10] Shi Jian-bo and Malik J. Normalized cuts and image segmentation[J]. IEEE Transaction on Pattern Analysis and Machine Intelligence, 2000, 22(8): 888-905.[11] Couprie C, Grady L, Najman L, and Talbot H. Power Watersheds: a new image segmentation framework extending graph cuts, random walker and optimal spanning forest[C]. ICCV09 , Kyoto, Japan, Sept. 29-Oct. 2, 2009: 731-738.[12] Cui Jing-yu, Yang Qiong, Wen Fang, Wu Qi-ying, Zhang Chang-shui, Van Cool L, and Tang Xiao-ou.Transductive object cutout[C], CVPR08, Anchorage, Alaska, USA, 23-28 June, 2008: 1-8.[13] Zhu Xiao-jing and Goldberg A B. Introduction to Semi-supervised Learning[M]. Synthesis Lectures on Artificial Intelligence and Machine Learning,3:1, USA. Morgan Claypool Publishers, 2009: 1-130. [14] Yin Pei, Criminisi A, Winn J, and Essa I. Bilayer segmentation of webcam videos using tree-based classifiers[J]. IEEE Transaction on Pattern Analysis and Machine Intelligence, 2010, 33(1): 30-42.[15] Kolmogorov V, Criminisi A, Blake A, Cross G, and Rother C. Bi-layer segmentation of binocular stereo video[C]. CVPR05, San Diego, CA, USA, 20-26 June, 2005: 407-414.[16] Quan Long, Wang Jing-dong, Tan Ping, and Yuan Lu. Image-based modeling by joint segmentation[J]. International Journal of Computer Vision, 2007, 75(1): 135-150.[17] Xiao Jianxiong, Wang Jingdong, Tan Ping, and Quan Long. Joint affinity propagation for multiple view segmentation[C]. ICCV07, Rio de Janeiro, Brazil, 14-20 October, 2007: 43-51.[18] Komodakis N, Paragios N, and Tzirita G. MRF optimization via dual decomposition: message-passing revisited[C]. ICCV07, Rio de Janeiro, Brazil, 14-20 October, 2007: 1-8.[19] Yang Jian-chao, Yu Kai, and Huang T. Supervised translation-invariant sparse coding[C]. CVPR10, San Francisco, USA, 13-18 June, 2010: 1-8.[20] Joachim T. Transductive learning via spectral graph partitioning[C]. International Conference on Machine Learning (ICML), Washington, DC, USA, 21-24 August, 2003: 290-297.[21] Felzenszwalb P F and Huttenlocher D P. Efficient Graph-based image segmentation[J]. International Journal of Computer Vision, 2004, 59(2): 167-181.

施引文献

资源附件(0)

访问统计