Combined mean shift and interactive multiple model for visual tracking by fusing multiple cues

Abstract

To overcome the tracking issues caused by the complex environment namely illumination variation and background clutters, tracking algorithm was proposed based on multi-cues fusion to construct a robust appearance model, indeed the traditional Mean Shift (MS) estimate the state associated with each sub appearance model, and the interactive multiple model (IMM) adjusts the weights of different cues and then combine the sub appearance models to estimate the general state. The proposed method is tested on public videos that present different environment issues. Experiences and comparisons conducted show the robustness of our methods in challenging tracking conditions.