Combined Kalman Filter and Multifeature Fusion Siamese Network for Real-Time Visual Tracking.

Lijun Zhou,Jianlin Zhang

doi:10.3390/s19092201

Abstract

SiamFC has a simple network structure and can be pretrained offline on a large data set, so it has attracted the attention of many researchers. It has no online learning process at all. Hence, there are no good solutions for some complex tracking scenarios such as occlusion and large target deformation. For this problem, we propose a method using the Kalman filter method and fusion multiresolution features and get multiple response scores. The Kalman filter acquires the target’s trajectory information, which is used to process complex tracking scenes and to change the selection method of the search area. This also enables our tracker to stably track fast moving targets.The introduction of the Kalman filter compensates for the shortcomings that SiamFC can only track offline, and the tracking network has an online learning process. The fusion of multiresolution features to obtain multiple response scores map helps the tracker to obtain robust features that can be adapted to a variety of tracking targets. Our proposed method has reached the state-of-the-art in testing on five data sets and can be run in real time (40 fps), including OTB2013, OTB2015, OTB50, VOT2015 and VOT 2016.

Highlights

Visual tracking aims to estimate the trajectory of a target in a video sequence
We propose introducing the Siamese network into the Kalman filtering method to obtain the target trajectory information so that the tracker can perform robust tracking on target occlusion, deformable, and fast motion scenes
The area under the curve (AUC) is calculated from the success plots where the average overlap precision (OP) of all videos is plotted within the threshold

Summary

Introduction

Visual tracking aims to estimate the trajectory of a target in a video sequence. It has wide applications ranging from human motion analysis, human–computer interaction to autonomous driving. The main difficulty of video tracking is how to use very limited training data (usually the bounding box in the first frame) to build a tracker that can adapt to various appearance changes, including scale variation, fast motion, occlusions, deformation, and background clutter. It should maintain both stable and real-time tracking. Others [9,10,11,12,13,14,15]

Methods

Results

Conclusion