Interacting Tracklets for Multi-Object Tracking

Lan, Long; Wang, Xinchao; Zhang, Shiliang; Tao, Dacheng; Gao, Wen; Huang, Thomas S.

doi:10.1109/tip.2018.2843129

Cited by 73 publications

(18 citation statements)

References 39 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…However, the motion model has proven helpful in multi-object tracking, which can help locate targets and realize the correspondence of multi-target labels in different frames. In most MOT applications [29,[39][40][41], a simple linear motion model is used to estimate the target state. Such motion models may cause a loss of tracking when the target turns quickly, suddenly stop or drives in reverse.…”

Section: Motion Modelmentioning

confidence: 99%

Spatial–Semantic and Temporal Attention Mechanism-Based Online Multi-Object Tracking

Meng

Wang

et al. 2020

Sensors

View full text Add to dashboard Cite

Multi-object tracking (MOT) plays a crucial role in various platforms. Occlusion and insertion among targets, complex backgrounds and higher real-time requirements increase the difficulty of MOT problems. Most state-of-the-art MOT approaches adopt the tracking-by-detection strategy, which relies on compute-intensive sliding windows or anchoring schemes to detect matching targets or candidates in each frame. In this work, we introduce a more efficient and effective spatial–temporal attention scheme to track multiple objects in various scenarios. Using a semantic-feature-based spatial attention mechanism and a novel Motion Model, we address the insertion and location of candidates. Some online-learned target-specific convolutional neural networks (CNNs) were used to estimate target occlusion and classify by adapting the appearance model. A temporal attention mechanism was adopted to update the online module by balancing current and history frames. Extensive experiments were performed on Karlsruhe Institute of Technologyand Toyota Technological Institute (KITTI) benchmarks and an Armored Target Tracking Dataset (ATTD) built for ground-armored targets. Experimental results show that the proposed method achieved outstanding tracking performance and met the actual application requirements.

show abstract

Section: Motion Modelmentioning

confidence: 99%

Spatial–Semantic and Temporal Attention Mechanism-Based Online Multi-Object Tracking

Meng

Wang

et al. 2020

Sensors

View full text Add to dashboard Cite

show abstract

“…Lan et al [14] propose an MOT approach that exploits interactions between tracklets. They introduce close and distant tracklet interaction.…”

Section: Related Workmentioning

confidence: 99%

Learn-select-track: An approach to multi-object tracking

Makhura

Woods

2019

Signal Processing: Image Communication

View full text Add to dashboard Cite

Object tracking algorithms rely on user input to learn the object of interest. In multi-object tracking, this can be a challenge when the user has to provide a lot of locations to track. This paper presents a new approach that reduces the need for user input in multi-tracking. The approach uses density based clustering to analyse the colours in one frame and find the best separation of colours. The colours selected from the detection are learned and used in subsequent frames to track the colours through the video. With this training approach, the user interaction is limited to selecting the colours rather than selecting the multiple location to be tracked. The training algorithm also provides online training even when training on thousands of features. Highlights Density based clustering for learning object features to track. Density based clustering for tracking objects in videos. Training and tacking without knowing the number of objects. Performance of the tacking and training algorithms.

show abstract

“…Recently, deep neural network (DNN) has been investigated intensively to learn the association cost function in a unified architecture combining both feature extraction and affinity metric [10,26,42]. Through training, the task and scenario prior can be automatically adapted by the candidate representation and estimation metric without manually tuning the hyper-parameters.…”

Section: Introductionmentioning

confidence: 99%

FAMNet: Joint Learning of Feature, Affinity and Multi-Dimensional Assignment for Online Multiple Object Tracking

Chu

Ling

2019

2019 IEEE/CVF International Conference on Computer Vision (ICCV)

240

108

View full text Add to dashboard Cite

Data association-based multiple object tracking (MOT) involves multiple separated modules processed or optimized differently, which results in complex method design and requires non-trivial tuning of parameters. In this paper, we present an end-to-end model, named FAMNet, where Feature extraction, Affinity estimation and Multi-dimensional assignment are refined in a single network. All layers in FAMNet are designed differentiable thus can be optimized jointly to learn the discriminative features and higher-order affinity model for robust MOT, which is supervised by the loss directly from the assignment ground truth. We also integrate single object tracking technique and a dedicated target management scheme into the FAMNet-based tracking system to further recover false negatives and inhibit noisy target candidates generated by the external detector. The proposed method is evaluated on a diverse set of benchmarks including MOT2015, MOT2017, KITTI-Car and UA-DETRAC, and achieves promising performance on all of them in comparison with state-of-the-arts.

show abstract

Interacting Tracklets for Multi-Object Tracking

Cited by 73 publications

References 39 publications

Spatial–Semantic and Temporal Attention Mechanism-Based Online Multi-Object Tracking

Spatial–Semantic and Temporal Attention Mechanism-Based Online Multi-Object Tracking

Learn-select-track: An approach to multi-object tracking

FAMNet: Joint Learning of Feature, Affinity and Multi-Dimensional Assignment for Online Multiple Object Tracking

Contact Info

Product

Resources

About