Abstract:Object tracking has remained a challenging problem in recent years. Most of the trackers can not work well, especially when dealing with problems such as similarly colored backgrounds, object occlusions, low illumination, or sudden illumination changes in real scenes. A centroid iteration algorithm using multiple features and a posterior probability criterion is presented to solve these problems. The model representation of the object and the similarity measure are two key factors that greatly influence the performance of the tracker. Firstly, this paper propose using a local texture feature which is a generalization of the local binary pattern (LBP) descriptor, which we call the double center-symmetric local binary pattern (DCS-LBP). This feature shows great discrimination between similar regions and high robustness to noise. By analyzing DCS-LBP patterns, a simplified DCS-LBP is used to improve the object texture model called the SDCS-LBP. The SDCS-LBP is able to describe the primitive structural information of the local image such as edges and corners. Then, the SDCS-LBP and the color are combined to generate the multiple features as the target model. Secondly, a posterior probability measure is introduced to reduce the rate of matching mistakes. Three strategies of target model update are employed. Experimental results show that our proposed algorithm is effective in improving tracking performance in complicated real scenarios compared with some state-of-the-art methods.

Object Detection Based On Multi-Viewpoint Histogram

Human Detection Method Based on Multi-Part Detector and Multi-Instance Learning

An HOG-CT Human Detector with Histogram-Based Search.

MVM3Det: A Novel Method for Multi-view Monocular 3D Detection

Object Detection Using Spatial Histogram Features

Novel Framework for Multi-view Object Detection through Combining Multiple Classifiers

Weighted Hough Voting For Multi-View Car Detection

Object Detection in Multispectral and Panchromatic Image Using Superpixel Segmentation and Multisource Feature

A Pedestrian-Detection Method Based on Heterogeneous Features and Ensemble of Multi-View–Pose Parts

Multiple View Object Recognition Based on Gabor and LIOP Feature

Multi-class Geospatial Object Detection and Geographic Image Classification Based on Collection of Part Detectors

Multiview Detection with Feature Perspective Transformation

Multi-view Pedestrian Detection Using Statistical Colour Matching

3D Object Detection Based on Multi-View Feature Point Matching

Object as Query: Equipping Any 2D Object Detector with 3D Detection Ability

A Novel Wide-Area Multiobject Detection System with High-Probability Region Searching

A Novel Image Copy Detection Scheme Based on the Local Multi-Resolution Histogram Descriptor

Object Tracking Using Local Multiple Features And A Posterior Probability Measure

Vision tracking algorithm under occlusion of multiple objects

Multi-view Visual Surveillance and Phantom Removal for Effective Pedestrian Detection.

Target Detection in Multiview and Multimodal Images Based on Markov Random Field