Abstract:Recently, combining multiple features into discriminative correlation filters to improve tracking representation has shown great potential in object tracking. Existing trackers apply fixed weights to fuse features or fuse response maps, which cannot adapt to the object drift well. Moreover, in the tracking algorithm, using cyclic shift to obtain training samples always cause boundary effect, resulting in dissatisfied tracking effect. Therefore, we first design a multiple features fusion method. Various handcrafted features are fused with the same weight, then the fused handcrafted features and deep features are fused by adaptive weights, which considerably improves the representation ability of the tracking object. Second, we propose a correlation filter object function model called Spatial-Channel Selection and Temporal Regularized Correlation Filters. We perform the grouping features selection from the dimensions of channel, spatial and temporal, so as to establish the relevance between the multi-channel features and the correlation filter. Finally, we transform the objective function of the model with equality constraint to augmented Lagrangian multiplier formula without constraint, which is divided into three subproblems with closed-form solutions. The optimal solution is obtained by iteratively solving three subproblems using Alternating Direction Multiplier Method (ADMM). We conduct extensive experiments in four public datasets, OTB-2013, OTB-2015, TC128, UAV123, and VOT2016. The experimental results represent our proposed tracker performs favorably against other prevailing trackers in success rate and precision.

Log-polar Mapping-Based Scale Space Tracking with Adaptive Target Response.

Collaborative Correlation Filters for Real-Time Tracking with Spatial Constraint.

A Scale Adaptive Kernel Correlation Filter Tracker With Feature Integration

High-speed Tracking with Multi-Templates Correlation Filters

Robust Estimation of Similarity Transformation for Visual Object Tracking

Scale-Adaptive Tracking Method by Combining Kernelized Correlation Filter with Geometric Estimation

Learning Scale-Adaptive Tight Correlation Filter for Object Tracking

Real-Time Scale Adaptive Visual Tracking with Context Information and Correlation Filters

Robust Scale Adaptive Visual Tracking With Correlation Filters

Scale-Adaptive Regression Position Prediction Tracking

Scale Adaptive Correlation Filtering Tracing Algorithm Based On Feature Fusion

Learning a Scale-and-rotation Correlation Filter for Robust Visual Tracking

Discriminative Scale Space Tracking

Multi-Scale Correlation Filtering Tracker Based on Adaptive Feature Selection

Combining Color Feature for Scale Adaptive Correlation Filter Tracker

Correlation filter tracking with adaptive proposal selection for accurate scale estimation

SCSTCF: Spatial-Channel Selection and Temporal Regularized Correlation Filters for visual tracking

Scale Adaptive Correlation Tracking Based on Convolutional Features

An Adaptive Multi-scale Tracking Method Based on Kernelized Correlation Filter.

Real-Time Scalable Visual Tracking Via Quadrangle Kernelized Correlation Filters.

Camouflaged Target Tracking Based on a Scaleadaptive Correlation Filter Algorithm