Kernel-based visual tracking with continuous adaptive distribution

The template updating problem of Kernel-based tracking (KBT) includes two aspects: target-scale update and target-model update. The proposed algorithm can update both tracking window's scale and target model by making use of continuous adaptive distribution. The ability of KBT can be complemented within its own framework with modest computation cost. The proposed tracking algorithm tries to get a balance between the stability of KBT and adaptability of CAMSHIFT for creating a robust tracker. c 2009 Society of Photo-Optical Instrumentation Engineers. [DOI: 10.1117/1.3125423]