Instance-aware representation learning and association for online multi-person tracking

Wu, Hefeng; Hu, Yafei; Wang, Keze; Li, Hanhui; Nie, Lin; Cheng, Hui

doi:10.1016/j.patcog.2019.04.018

Cited by 27 publications

(7 citation statements)

References 44 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…There are also some works focusing on the instance-aware tracking in multi-person tracking task. Wu et al [55] learn instance-aware representations of the tracked person by a multibranch neural network where each branch (instance-subnet) corresponds to an individual. Differently, our approach obtains instance-aware representation by directly classifying every target as a specific category against plenty of the other samples.…”

Section: Instance-aware Trackingmentioning

confidence: 99%

Explicitly Modeling the Discriminability for Instance-Aware Visual Object Tracking

Wang¹,

Yang²,

Liu³

2021

Preprint

View full text Add to dashboard Cite

Visual object tracking performance has been dramatically improved in recent years, but some severe challenges remain open, like distractors and occlusions. We suspect the reason is that the feature representations of the tracking targets are only expressively learned but not fully discriminatively modeled. In this paper, we propose a novel Instance-Aware Tracker (IAT) to explicitly excavate the discriminability of feature representations, which improves the classical visual tracking pipeline with an instance-level classifier. First, we introduce a contrastive learning mechanism to formulate the classification task, ensuring that every training sample could be uniquely modeled and be highly distinguishable from plenty of other samples. Besides, we design an effective negative sample selection scheme to contain various intra and inter classes in the instance classification branch. Furthermore, we implement two variants of the proposed IAT, including a video-level one and an object-level one. They realize the concept of instance in different granularity as videos and target bounding boxes, respectively. The former enhances the ability to recognize the target from the background while the latter boosts the discriminative power for mitigating the target-distractor dilemma. Extensive experimental evaluations on 8 benchmark datasets show that both two versions of the proposed IAT achieve leading results against state-of-the-art methods while running at 30FPS. Code will be available when it is published.

show abstract

Section: Instance-aware Trackingmentioning

confidence: 99%

Explicitly Modeling the Discriminability for Instance-Aware Visual Object Tracking

Wang¹,

Yang²,

Liu³

2021

Preprint

View full text Add to dashboard Cite

show abstract

“…Sadeghian et al [9] applied a recurrent neural network (RNN) in modeling appearance, motion, and interaction information of objects to compute their similarity to detections. Wu et al [10] designed a multi-branch neural network to predict the confidence and location of objects. Yoon et al [11] exploited the one-shot learning MOT framework based on an attention mechanism.…”

Section: Of 16mentioning

confidence: 99%

Hierarchical-Matching-Based Online and Real-Time Multi-Object Tracking with Deep Appearance Features

2020

Algorithms

View full text Add to dashboard Cite

Based on tracking-by-detection, we propose a hierarchical-matching-based online and real-time multi-object tracking approach with deep appearance features, which can effectively reduce the false positives (FP) in tracking. For the purpose of increasing the accuracy rate of data association, we define the trajectory confidence using its position information, appearance information, and the information of historical relevant detections, after which we can classify the trajectories into different levels. In order to obtain discriminative appearance features, we developed a deep convolutional neural network to extract the appearance features of objects and trained it on a large-scale pedestrian re-identification dataset. Last but not least, we used the proposed diverse and hierarchical matching strategy to associate detection and trajectory sets. Experimental results on the MOT benchmark dataset show that our proposed approach performs well against other online methods, especially for the metrics of FP and frames per second (FPS).

show abstract

“…The main goal of tracking multiple objects is to discover correspondences and carries out matching among different objects in adjoining frames, which remained a matching issue in several visual applications. The outcomes of object detection 18 are the major data cues for tracking multiple objects 19 . Hence, one major issue of tracking multiple objects relies on managing noisy detection.…”

Section: Introductionmentioning

confidence: 99%

“…The outcomes of object detection 18 are the major data cues for tracking multiple objects. 19 Hence, one major issue of tracking multiple objects relies on managing noisy detection. The majority of classical multi-object tracking techniques concentrated on devising a proper object relationship approach in a way that the objects in various frames are coordinated in an optimal manner using the cost function.…”

Section: Introductionmentioning

confidence: 99%

Object detection and multi‐object tracking based on optimized deep convolutional neural network and unscented Kalman filtering

Jaganathan¹,

Panneerselvam

Kumaraswamy³

2022

Concurrency and Computation

View full text Add to dashboard Cite

Summary The tracking of multiple objects has gained immense interest in several applications that ranges from video surveillance and robotics. The goal of tracking multiple objects is to discover correspondences and perform matching among various objects in frames. However, occlusion is a major issue in several classical techniques, which fail to track multiple moving objects. Here, the extraction of frames is done to obtain information about an action. Thereafter, the segmentation of objects is done using the entropy weighting K‐means (EWKM) algorithm to generate optimum segments from the frame. After that, the detection of the object is done using a deep convolutional neural network (DCNN) to obtain objects from segments. Proposed Jaya‐political optimizer (Jaya‐PO) is used to train DCNN, which is devised by collaborating political optimizer (PO) and Jaya optimization algorithm. After detecting the object, the tracking of multiple objects is done using the proposed Henry gas solubility optimized unscented Kalman filtering (HGSO‐based UKF), wherein the UKF is modified with HGSO. The proposed Jaya‐PO‐based DCNN and HGSO‐based UKF outperformed other methods with high accuracy of 92%, high multiple object tracking precision (MOTP) of 87.6%, high sensitivity of 91.7%, and high specificity of 92%.

show abstract

Instance-aware representation learning and association for online multi-person tracking

Cited by 27 publications

References 44 publications

Explicitly Modeling the Discriminability for Instance-Aware Visual Object Tracking

Explicitly Modeling the Discriminability for Instance-Aware Visual Object Tracking

Hierarchical-Matching-Based Online and Real-Time Multi-Object Tracking with Deep Appearance Features

Object detection and multi‐object tracking based on optimized deep convolutional neural network and unscented Kalman filtering

Contact Info

Product

Resources

About