Scalable Kernel Correlation Filter With Sparse Feature Integration

Andres Solis Montero, Jochen Lang, Robert Laganiere; The IEEE International Conference on Computer Vision (ICCV) Workshops, 2015, pp. 24-31

Abstract


Correlation filters for long-term visual object tracking have recently seen great interest. Although they present competitive performance results, there is still a need for im- proving their tracking capabilities. In this paper, we present a fast scalable solution based on the Kernalized Correlation Filter (KCF) framework. We introduce an adjustable Gaus- sian window function and a keypoint-based model for scale estimation to deal with the fixed size limitation in the Ker- nelized Correlation Filter. Furthermore, we integrate the fast HoG descriptors and Intel's Complex Conjugate Sym- metric (CCS) packed format to boost the achievable frame rates. We test our solution using the Visual Tracker Bench- mark and the VOT Challenge datasets. We evaluate our tracker in terms of precision and success rate, accuracy, robustness and speed. The empirical evaluations demon- strate clear improvements by the proposed tracker over the KCF algorithm while ranking among the top state-of-the- art trackers.

Related Material


[pdf]
[bibtex]
@InProceedings{Montero_2015_ICCV_Workshops,
author = {Solis Montero, Andres and Lang, Jochen and Laganiere, Robert},
title = {Scalable Kernel Correlation Filter With Sparse Feature Integration},
booktitle = {The IEEE International Conference on Computer Vision (ICCV) Workshops},
month = {December},
year = {2015}
}