Learning Hybrid Template by EM-Type Algorithm

摘要：

The article proposes to improve the active basis model by incorporating both unaligned training examples and nonalignable sketches in images. EM-type algorithm 8 can learn the objects appear at unknown orientations, locations and scales in the training images. And non-alignable sketches 11 can be summarized in average sketches over the image lattice. This article proposes to add the score of the non-alignable sketches to the likelihood of M-step, so the learned active basis model by EM-type algorithm should be more accurate. Our experiments show that the proposed model can achieve considerable improvement in ROC for most of object categories.

作者: Bin Lai Deng-Yi Zhang Cheng-Zhang Qu Jian-Hui Zhao Zhi-Yong Yuan

作者单位: Computer School Wuhan University, Wuhan, China

会议类型: 国际会议

会议名称: The First International Conference on Multimedia Information Networking and Security(第一届国际多媒体网络信息安全会议 MINES 2009)

会议地点: 武汉

会议语种:英文

页码: 629-632

在线出版日期: 2009-11-18（万方平台首次上网日期，不代表论文的发表时间）