Robust online multiple object tracking based on the confidence-based relative motion network and correlation filter

Park, Sehoon; Lee, Kyuewang; Yoon, Kuk-Jin

doi:10.1109/icip.2016.7533007

Cited by 6 publications

(6 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Discriminative correlation filters (DCFs) have been successfully exploited in multi-target tracking applications due to their high computational efficiency. For example, integrate correlation filters (ICFs) uses a confidence-based relative motion network to perform a two-step data association, where CFs are employed as a verifying step to confirm the target estimation [15]. Yang et al [16] and [17] use multiple single object trackers based on the kernelized correlation filters (KCFs) in parallel for fast tracking.…”

Section: Multi-target Trackingmentioning

confidence: 99%

Quick Overview for Multi-target Tracking

Ben¹

2020

Preprint

View full text Add to dashboard Cite

Tracking an unknown and time-varying number of targets (e.g., speakers) in indoor environments using audio-visual (AV) modalities has received increasing interest in numerous fields including video conferencing, individual speaker discrimination, and human-computer interaction. The audio-visual sequential Monte Carlo probability hypothesis density (AV-SMC-PHD) filter is a popular baseline for multi-target tracking, offering an elegant framework for fusing audio-visual information and dealing with a varying number of speakers. However, the performance of this filter can be adversely affected by the weight degeneracy problem, where the weights of most of the particles may become very small, while only few remain significant, during the iteration of the algorithm. In this paper, we will short discuss the multi-target tracking.

show abstract

Section: Multi-target Trackingmentioning

confidence: 99%

Quick Overview for Multi-target Tracking

Ben¹

2020

Preprint

View full text Add to dashboard Cite

show abstract

“…Both batch methods [20,21] and online methods [22,23] explore how to learn a similarity function for data association. More recent studies on MOT have integrated hierarchical features from deep convolution networks [24,25] and correlation filters [26]. In addition, use of a reinforcement learning algorithm has been proposed to link data in online MOT: e.g., Markov decision processes (MDP) have proved suitable for dynamic environments [27].…”

Section: Related Workmentioning

confidence: 99%

Fluid-inspired field representation for risk assessment in road scenes

Zhu

Xue

et al. 2020

Comp. Visual Media

View full text Add to dashboard Cite

Prediction of the likely evolution of traffic scenes is a challenging task because of high uncertainties from sensing technology and the dynamic environment. It leads to failure of motion planning for intelligent agents like autonomous vehicles. In this paper, we propose a fluid-inspired model to estimate collision risk in road scenes. Multi-object states are detected and tracked, and then a stable fluid model is adopted to construct the risk field. Objects’ state spaces are used as the boundary conditions in the simulation of advection and diffusion processes. We have evaluated our approach on the public KITTI dataset; our model can provide predictions in the cases of misdetection and tracking error caused by occlusion. It proves a promising approach for collision risk assessment in road scenes.

show abstract

“…r denote the mean value and the standard deviation of the sidelobes. Similar works [35] and [20] used the PSRs as a gating technique to confirm the predicted state or detect tracking failures. However, our work focuses on enhancing the association step.…”

Section: B Discriminative Correlation Matchingmentioning

confidence: 99%

“…To investigate how correlation filters can improve multiple human tracking, [33] and [34] have been proposed to apply multiple single object trackers based on the Kernelized Correlation Filters (KCFs) in parallel for fast tracking. In [35], authors proposed to integrate correlation filters (CFs) and a confidence-based relative motion network to perform a two-step data association to track multiple objects, where CFs are employed as a verifying step to confirm the target estimates. Furthermore, a recent RFS based tracking approach [20] was proposed to perform the KCF as an extended step after the PHD update, where the KCF is mainly used to perform the refinement of target prediction oriented by the label tree technique [36].…”

Section: Introductionmentioning

confidence: 99%

Multi-Level Cooperative Fusion of GM-PHD Filters for Online Multiple Human Tracking

Angelini

Chambers

et al. 2019

IEEE Trans. Multimedia

View full text Add to dashboard Cite

In this paper, we propose a multi-level cooperative fusion approach to address the online multiple human tracking problem in a Gaussian Mixture Probability Hypothesis Density (GM-PHD) filter framework. The proposed fusion approach consists essentially of three steps. Firstly, we integrate two human detectors with different characteristics (full-body and bodyparts), and investigate their complementary benefits for tracking multiple targets. For each detector domain, we then propose a novel Discriminative Correlation Matching (DCM) model, and fuse it with spatio-temporal information to address ambiguous identity association in the GM-PHD filter. Finally, we develop a robust fusion center with virtual and real zones to make a global decision based on preliminary candidate targets generated by each detector. This center also mitigates the sensitivity of missed detections in the Generalized Covariance Intersection (GCI) fusion process, thereby improving the fusion performance and tracking consistency. Experiments on the MOTChallenge Benchmark demonstrate the proposed method achieves improved performance over other state-of-the-art RFS based tracking methods.

show abstract

Robust online multiple object tracking based on the confidence-based relative motion network and correlation filter

Cited by 6 publications

References 17 publications

Quick Overview for Multi-target Tracking

Quick Overview for Multi-target Tracking

Fluid-inspired field representation for risk assessment in road scenes

Multi-Level Cooperative Fusion of GM-PHD Filters for Online Multiple Human Tracking

Contact Info

Product

Resources

About