Online multi-object tracking via robust collaborative model and sample selection

Naiel, Mohamed A.; Ahmad, M. Omair; Swamy, M.N.S.; Lim, Jongwoo; Yang, Ming–Hsuan

doi:10.1016/j.cviu.2016.07.003

Cited by 28 publications

(23 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Similar to the above work, Lan et al [35] also adopt tracking-by-detection mechanism but with a more sophisticated tracker update routine. Naiel et al [36] try to decrease the false negative rate (miss detection caused by a simple detector) of the previous pipeline without reducing speed.…”

Section: B Online Tracking 1) Hand-crafted Featuresmentioning

confidence: 96%

The New High-Performance Face Tracking System based on Detection-Tracking and Tracklet-Tracklet Association in Semi-Online Mode

Ly¹,

Nguyen²,

Vong³

et al. 2020

IJACSA

View full text Add to dashboard Cite

Despite recent advances in multiple object tracking and pedestrian tracking, multiple-face tracking remains a challenging problem. In this work, the authors propose a framework to solve the problem in semi-online manner (the framework runs in real-time speed with two-second delay). The proposed framework consists of two stages: detection-tracking and tracklet-tracklet association. Detection-tracking stage is for creating short tracklets. Tracklet-tracklet association is for merging and assigning identifications to those tracklets. To the best of the authors' knowledge, the authors make contributions in three aspects: 1) the authors adopt a principle often used in online approaches as a part of the framework and introduce a tracklet-tracklet association stage to leverage future information; 2) the authors propose a motion affinity metric to compare trajectories of two tracklets; 3) the authors propose an efficient way to employ deep features in comparing tracklets of faces. The authors achieved 78.7% precision plot AUC, 68.1% success plot AUC on MobiFace dataset (test set). On OTB dataset, the authors achieved 78.2% and 72.5% precision plot AUC, 51.9% and 43.9% success plot AUC on normal and difficult face subsets, respectively. The average speed was maintained at around 44 FPS. In comparison to the state-of-the-art methods, the proposed framework's performance maintains high rankings in top 3 on two datasets while keeping the processing speed higher than the other methods in top 3.

show abstract

Section: B Online Tracking 1) Hand-crafted Featuresmentioning

confidence: 96%

The New High-Performance Face Tracking System based on Detection-Tracking and Tracklet-Tracklet Association in Semi-Online Mode

Ly¹,

Nguyen²,

Vong³

et al. 2020

IJACSA

View full text Add to dashboard Cite

show abstract

“…Moving object detection is one of the hotspots in machine vision research (Zhang et al, 2014(Zhang et al, , 2016Choi and Maurer, 2016;Naiel et al, 2017). Fischler and Bolles (1981) proposed a paradigm for model fitting with applications to image analysis and automated cartography.…”

Section: Related Workmentioning

confidence: 99%

Walking Human Detection Using Stereo Camera Based on Feature Classification Algorithm of Second Re-projection Error

et al. 2019

View full text Add to dashboard Cite

This paper presents a feature classification method based on vision sensor in dynamic environment. Aiming at the detected targets, a double-projection error based on orb and surf is proposed, which combines texture constraints and region constraints to achieve accurate feature classification in four different environments. For dynamic targets with different velocities, the proposed classification framework can effectively reduce the impact of large-area moving targets. The algorithm can classify static and dynamic feature objects and optimize the conversion relationship between frames only through visual sensors. The experimental results show that the proposed algorithm is superior to other algorithms in both static and dynamic environments.

show abstract

“…Many state-of-the-art people trackers 5,[14][15][16][17][18][19][20][21]42 use the PETS2009 S2.L1 video sequence to evaluate their work in terms of aforementioned MOTA and MOTP matrices. In the computation of MOTA and MOTP, all trackers use the same ground truth, 39 except in Refs.…”

Section: Performance Comparisonmentioning

confidence: 99%

“…13 The evaluation results show that deploying our occlusion handling on the baseline tracker 10 considerably improves the accuracy. On the S2.L1 video of the PETS2009 dataset, which is widely used as a benchmark in literature, our tracker outperforms state-ofthe-art techniques 5,[14][15][16][17][18][19][20][21] in terms of multiple object tracking accuracy (MOTA).…”

Section: Introductionmentioning

confidence: 99%

Occlusion handling framework for tracking in smart camera networks by per-target assistance task assignment

Deboeverie

Veelaert

et al. 2017

J. Electron. Imaging

View full text Add to dashboard Cite

Abstract. Occlusion is one of the most difficult challenges in the area of visual tracking. We propose an occlusion handling framework to improve the performance of local tracking in a smart camera view in a multicamera network. We formulate an extensible energy function to quantify the quality of a camera's observation of a particular target by taking into account both person-person and object-person occlusion. Using this energy function, a smart camera assesses the quality of observations over all targets being tracked. When it cannot adequately observe of a target, a smart camera estimates the quality of observation of the target from view points of other assisting cameras. If a camera with better observation of the target is found, the tracking task of the target is carried out with the assistance of that camera. In our framework, only positions of persons being tracked are exchanged between smart cameras. Thus, communication bandwidth requirement is very low. Performance evaluation of our method on challenging video sequences with frequent and severe occlusions shows that the accuracy of a baseline tracker is considerably improved. We also report the performance comparison to the state-of-the-art trackers in which our method outperforms.

show abstract

Online multi-object tracking via robust collaborative model and sample selection

Cited by 28 publications

References 17 publications

The New High-Performance Face Tracking System based on Detection-Tracking and Tracklet-Tracklet Association in Semi-Online Mode

The New High-Performance Face Tracking System based on Detection-Tracking and Tracklet-Tracklet Association in Semi-Online Mode

Walking Human Detection Using Stereo Camera Based on Feature Classification Algorithm of Second Re-projection Error

Occlusion handling framework for tracking in smart camera networks by per-target assistance task assignment

Contact Info

Product

Resources

About