Feature-driven motion model-based particle-filter tracking method with abrupt motion handling

Abstract

The potential for the research of object tracking in computer vision has been well established, but previous object-tracking methods, which consider only continuous and smooth motion, are limited in handling abrupt motions. We introduce an efficient algorithm to tackle this limitation. A feature-driven (FD) motion model-based features from accelerated segment test (FAST) feature matching is proposed in the particle-filtering framework. Various evaluations have demonstrated that this motion model can improve existing methods' performances to handle abrupt motion significantly. The proposed model can be applied to most existing particlefilter tracking methods.

Position and locality constrained soft coding for human action recognition

“…To extract ROI, we label a bounding box for actor at the first frame in each split and then track actor with the object tracking algorithm proposed by Liu et al 44 to obtain the ROI for KTH dataset; the annotation bounding boxes are used for extracting ROI for UCF sports dataset; the mask labeled results from Ref. 45 are used for Weizmann dataset.…”

Refined particle swarm intelligence method for abrupt motion tracking

“…While considerable research efforts exist in relation to visual tracking, only a handful corresponds to abrupt motion [23,14,24,16]. Abrupt motion can be defined as situations where the objects motion changes at adjacent frames with unknown pattern in scenarios such as i) partially low-frame rate, ii) switching of cameras view in a topology network or iii) the irregular motion of the object.…”