Recurrent Autoregressive Networks for Online Multi-object Tracking

Fang, Kuan; Xiang, Yu; Li, Xiaocheng; Savarese, Silvio

doi:10.1109/wacv.2018.00057

Cited by 254 publications

(118 citation statements)

References 31 publications

Supporting

Mentioning

117

Contrasting

Order By: Relevance

“…Table 7 compares our dual L 1 normalized context-aware tensor power iteration method with the state-of the art methods on the MOT16 challenge benchmark dataset where "Ours-1" refers to the results of our method using two sets of parameter values and "Ours-2" refers to the results of our method using one set of parameter values for all the sequences. It is seen that our tracker is a strong competitor to the competing online trackers (Yu et al 2016;Wojke et al 2017;Bewley et al 2016;Bochinski et al 2017;Fang et al 2018) which are pairwise association-based. In particular, our method returns the highest identified detection score and MT, and fewer fragments among all the online pairwise association-based methods while maintaining competitive MOTA scores, ML, and identity switches.…”

Section: Mot16 Challenge Benchmarkmentioning

confidence: 98%

Dual L1-Normalized Context Aware Tensor Power Iteration and Its Applications to Multi-object Tracking and Multi-graph Matching

Shi

Zhou

et al. 2019

Int J Comput Vis

View full text Add to dashboard Cite

The multi-dimensional assignment problem is universal for data association analysis such as data association-based visual multi-object tracking and multi-graph matching. In this paper, multi-dimensional assignment is formulated as a rank-1 tensor approximation problem. A dual L 1-normalized context/hyper-context aware tensor power iteration optimization method is proposed. The method is applied to multi-object tracking and multi-graph matching. In the optimization method, tensor power iteration with the dual unit norm enables the capture of information across multiple sample sets. Interactions between sample associations are modeled as contexts or hyper-contexts which are combined with the global affinity into a unified optimization. The optimization is flexible for accommodating various types of contextual models. In multi-object tracking, the global affinity is defined according to the appearance similarity between objects detected in different frames. Interactions between objects are modeled as motion contexts which are encoded into the global association optimization. The tracking method integrates high order motion information and high order appearance variation. The multi-graph matching method carries out matching over graph vertices and structure matching over graph edges simultaneously. The matching consistency across multi-graphs is based on the high-order tensor optimization. Various types of vertex affinities and edge/hyper-edge affinities are flexibly integrated. Experiments on several public datasets, such as the MOT16 challenge benchmark, validate the effectiveness of the proposed methods. Keywords Multi-dimensional assignment • Context/hyper-context aware tensor power iteration • Multi-object tracking • Multi-graph matching Communicated by M. Hebert.

show abstract

Section: Mot16 Challenge Benchmarkmentioning

confidence: 98%

Dual L1-Normalized Context Aware Tensor Power Iteration and Its Applications to Multi-object Tracking and Multi-graph Matching

Shi

Zhou

et al. 2019

Int J Comput Vis

View full text Add to dashboard Cite

show abstract

“…In [12], authors propose a temporal generative network namely recurrent autoregressive network to model the appearance and motion features in temporal sequences. It strongly couples internal and external memory with the net-work thus incorporating information about previous frames trajectories and long term dependencies.…”

Section: Multi-object Trackingmentioning

confidence: 99%

Aerial Multi-Object Tracking by Detection Using Deep Association Networks

Jadhav

Mukherjee

Kaushik

et al. 2020

2020 National Conference on Communications (NCC)

View full text Add to dashboard Cite

A lot a research is focused on object detection and it has achieved significant advances with deep learning techniques in recent years. Inspite of the existing research, these algorithms are not usually optimal for dealing with sequences or images captured by drone-based platforms, due to various challenges such as view point change, scales, density of object distribution and occlusion. In this paper, we develop a model for detection of objects in drone images using the VisDrone2019 DET dataset. Using the RetinaNet model as our base, we modify the anchor scales to better handle the detection of dense distribution and small size of the objects. We explicitly model the channel interdependencies by using Squeeze-and-Excitation (SE) blocks that adaptively recalibrates channel-wise feature responses. This helps to bring significant improvements in performance at a slight additional computational cost. Using this architecture for object detection, we build a custom DeepSORT network for object detection on the VisDrone2019 MOT dataset by training a custom Deep Association network for the algorithm.

show abstract

“…This tracker also conducts a data association immediately after the prediction phase, eliminating the need for computationally expensive labeling procedures such as clustering. Fang et al [27] proposed a recurrent autoregressive network, which is a temporal generative modeling framework used to characterize the appearance and motion dynamics of multiple objects over time. The external memory explicitly stores previous inputs of each trajectory in a time window, whereas the internal memory learns to summarize the long-term tracking history and associate detections by processing the external memory.…”

Section: Related Studiesmentioning

confidence: 99%

Online Multiple Object Tracking Using Rule Distillated Siamese Random Forest

Lee

Kim

2020

IEEE Access

View full text Add to dashboard Cite

In a multiple object tracking (MOT) system, an association check between the tracker and detected objects is an important factor in determining the tracking performance. Siamese convolution neural network (CNN) is the most popular data association method in MOT owing to its good matching performance and network sharing support. However, it is unsuitable for real-time online tracking in low-end systems because numerous parameters and operations are still required. In this paper, instead of a CNN, we propose using a SiameseRF algorithm which combines Siamese structure and random forest (RF), enabling highspeed learning and classification. SiameseRF has a shared-rule based Siamese structure rather than shared weight, which improves the matching performance and solves existing slow CNN-based tracking issues. During the learning process, the shared RFs consisting of tree rules are learned in the directions of increasing similarity to the positive pair {anchor, positive} and increasing difference between the negative pair {anchor, negative}. However, because many rules that make up SiameseRF remain a burden for online processing, this study proposes an additional rule distillation algorithm to effectively remove redundant and unimportant rules causing an overfitting with SiameseRF. This reduction in the number of rules reduces the processing time and number of parameters in the rule distilled SiameseRF. In experiments conducted on MOT benchmark datasets, our proposed 30% rule distilled SiameseRF achieved up to a 1.12-times faster speed and a 1.13-times higher compression rate than basic SiameseRF while maintaining a similar or somewhat better tracking performance than other state-of-the-art CNN-based MOT algorithms.

show abstract

Recurrent Autoregressive Networks for Online Multi-object Tracking

Cited by 254 publications

References 31 publications

Dual L1-Normalized Context Aware Tensor Power Iteration and Its Applications to Multi-object Tracking and Multi-graph Matching

Dual L1-Normalized Context Aware Tensor Power Iteration and Its Applications to Multi-object Tracking and Multi-graph Matching

Aerial Multi-Object Tracking by Detection Using Deep Association Networks

Online Multiple Object Tracking Using Rule Distillated Siamese Random Forest

Contact Info

Product

Resources

About