Learning structured visual dictionary for object tracking

作者：

Highlights：

• A tracking algorithm which exploits both appearance and geometric information

• Two complementary features are adopted in building the appearance dictionary.

• A shape context approach to capture the stable geometric patterns of keypoints

• The proposed tracking algorithm performs well in challenging conditions.

摘要

•A tracking algorithm which exploits both appearance and geometric information•Two complementary features are adopted in building the appearance dictionary.•A shape context approach to capture the stable geometric patterns of keypoints•The proposed tracking algorithm performs well in challenging conditions.

论文关键词：Object tracking,Bag of features,Appearance model,Geometric relationship

论文评审过程：Received 28 December 2012, Revised 12 August 2013, Accepted 23 September 2013, Available online 24 October 2013.

论文官网地址：https://doi.org/10.1016/j.imavis.2013.09.008