Discriminative Scale Space Tracking

Abstract

Accurate scale estimation of a target is a challenging research problem in visual object tracking. Most state-of-the-art methods employ an exhaustive scale search to estimate the target size. The exhaustive search strategy is computationally expensive and struggles when encountered with large scale variations. This paper investigates the problem of accurate and robust scale estimation in a tracking-by-detection framework. We propose a novel scale adaptive tracking approach by learning separate discriminative correlation filters for translation and scale estimation. The explicit scale filter is learned online using the target appearance sampled at a set of different scales. Contrary to standard approaches, our method directly learns the appearance change induced by variations in the target scale. Additionally, we investigate strategies to reduce the computational cost of our approach. Extensive experiments are performed on the OTB and the VOT2014 datasets. Compared to the standard exhaustive scale search, our approach achieves a gain of 2.5 percent in average overlap precision on the OTB dataset. Additionally, our method is computationally efficient, operating at a 50 percent higher frame rate compared to the exhaustive scale search. Our method obtains the top rank in performance by outperforming 19 state-of-the-art trackers on OTB and 37 state-of-the-art trackers on VOT2014.

Video Based Accurate Step Counting for Treadmills

“…At present, there are many algorithm can be used to track human motion accurately in real time. In order to accurate and robust scale estimation for real-time visual tracking, Martin et al [8] proposed a novel scale adaptive tracking approach by learning separate discriminative correlation filters for translation and scale estimation, which can accurately tracking the position of shoulder in the processing of running on the treadmill. In this paper, we adopt DSST algorithm for robust shoulder tracking.…”

A comparison study on the adaptive scale estimation of correlation filter-based visual tracking methods

“…And these correlation filterbased trackers are susceptible to drifting. Danelljan et al [26] employed an adaptive feature dimensionality reduction method as in Ref. [27] to reduce the computational cost, while tracking the performance is preserved.…”

“…A collaborative correlation tracker is proposed in [28]. The experiment results [14,24] show that by combining the scale estimation with translation filter, their approaches outperform 19 state-of-the-art trackers in the OTB dataset.…”

“…In this section, we first briefly introduce the basic correlation filter-based visual tracking method; then, three typical strategies for scale estimation are presented, that is, multi-resolution-based [29], joint scale space filter, iterative joint scale space [14,24].…”

“…For each Fig. 1 Architecture of joint scale space estimation method [24] an image patch I n of size a n P × a n R centered around the target. Here, a denotes the scale factor between feature layers.…”