Integrating tracking with fine object segmentation
作者：
Highlights：
• A novel method for joint tracking and fine object segmentation in videos
• Efficient integration of EM-based tracking and Random Walker-based image segmentation
• No strong constraints are imposed on the target appearance or the camera motion.
• Experimental evaluation against a large collection of state-of-the-art methods
• Explicit and efficient fine object segmentation facilitates drift-free tracking.
摘要
•A novel method for joint tracking and fine object segmentation in videos•Efficient integration of EM-based tracking and Random Walker-based image segmentation•No strong constraints are imposed on the target appearance or the camera motion.•Experimental evaluation against a large collection of state-of-the-art methods•Explicit and efficient fine object segmentation facilitates drift-free tracking.
论文关键词：Kernel-based object tracking,Video-based object segmentation,Tracking by segmentation,Bayesian Inference,Data fusion,Expectation Maximization,Random Walker
论文评审过程：Received 7 December 2012, Revised 10 June 2013, Accepted 18 July 2013, Available online 7 August 2013.
论文官网地址：https://doi.org/10.1016/j.imavis.2013.07.008
