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

Hu, Weiming; Shi, Xinchu; Zhou, Zongwei; Ling, Haibin; Maybank, Stephen J.

doi:10.1007/s11263-019-01231-y

Cited by 11 publications

(4 citation statements)

References 58 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…There has been a large body of research on single-camera MOT. These methods focus on the data association step, for which the (lifted) multicut problem [37,38,39], the lifted disjoint paths problem [21], maximum clique [10,47], multigraph-matching [22], binary quadratic optimization [17,18,19,40] was used. Another area is building end-to-end differentiable frameworks for both detector and data association [1,9,42,49,52].…”

Section: Related Workmentioning

confidence: 99%

LMGP: Lifted Multicut Meets Geometry Projections for Multi-Camera Multi-Object Tracking

Nguyen¹,

Henschel²,

Rosenhahn³

et al. 2021

Preprint

View full text Add to dashboard Cite

Multi-Camera Multi-Object Tracking is currently drawing attention in the computer vision field due to its superior performance in real-world applications such as video surveillance with crowded scenes or in vast space. In this work, we propose a mathematically elegant multi-camera multiple object tracking approach based on a spatialtemporal lifted multicut formulation. Our model utilizes state-of-the-art tracklets produced by single-camera trackers as proposals. As these tracklets may contain ID-Switch errors, we refine them through a novel pre-clustering obtained from 3D geometry projections. As a result, we derive a better tracking graph without ID switches and more precise affinity costs for the data association phase. Tracklets are then matched to multi-camera trajectories by solving a global lifted multicut formulation that incorporates short and long-range temporal interactions on tracklets located in the same camera as well as inter-camera ones. Experimental results on the WildTrack dataset yield near-perfect result, outperforming state-of-the-art trackers on Campus while being on par on the PETS-09 dataset. We will make our implementations available upon acceptance of the paper 1 .

show abstract

Section: Related Workmentioning

confidence: 99%

LMGP: Lifted Multicut Meets Geometry Projections for Multi-Camera Multi-Object Tracking

Nguyen¹,

Henschel²,

Rosenhahn³

et al. 2021

Preprint

View full text Add to dashboard Cite

show abstract

“…In MOT task, the application of graph matching is very limited. To the best of our knowledge, [20] is the first to formulate the MOT task as a graph matching problem and use dual L1-normalized tensor power iteration method to solve it. Different from [20] that directly extracts the features from an off-the-shelf neural network, we propose to guide the feature learning by the optimization problem, which can both enjoy the power of deep feature learning and combinatorial optimization.…”

Section: Related Workmentioning

confidence: 99%

“…To the best of our knowledge, [20] is the first to formulate the MOT task as a graph matching problem and use dual L1-normalized tensor power iteration method to solve it. Different from [20] that directly extracts the features from an off-the-shelf neural network, we propose to guide the feature learning by the optimization problem, which can both enjoy the power of deep feature learning and combinatorial optimization. This joint training manner of representation and optimization problem also eliminate the inconsistencies between the training and inference.…”

Section: Related Workmentioning

confidence: 99%

Learnable Graph Matching: Incorporating Graph Partitioning with Deep Feature Learning for Multiple Object Tracking

He¹,

Huang²,

Wang³

et al. 2021

Preprint

View full text Add to dashboard Cite

Data association across frames is at the core of Multiple Object Tracking (MOT) task. This problem is usually solved by a traditional graph-based optimization or directly learned via deep learning. Despite their popularity, we find some points worth studying in current paradigm: 1) Existing methods mostly ignore the context information among tracklets and intra-frame detections, which makes the tracker hard to survive in challenging cases like severe occlusion.2) The end-to-end association methods solely rely on the data fitting power of deep neural networks, while they hardly utilize the advantage of optimizationbased assignment methods. 3) The graph-based optimization methods mostly utilize a separate neural network to extract features, which brings the inconsistency between training and inference. Therefore, in this paper we propose a novel learnable graph matching method to address these issues. Briefly speaking, we model the relationships between tracklets and the intra-frame detections as a general undirected graph. Then the association problem turns into a general graph matching between tracklet graph and detection graph. Furthermore, to make the optimization end-to-end differentiable, we relax the original graph matching into continuous quadratic programming and then incorporate the training of it into a deep graph network with the help of the implicit function theorem. Lastly, our method GMTracker, achieves state-ofthe-art performance on several standard MOT datasets. Our code will be available at https://github.com/ jiaweihe1996/GMTracker.

show abstract

“…Related Work. Other problems have also been used to model the object association task in MOT, such as CLIQUE (Zamir, Dehghan, and Shah 2012;Dehghan, Assari, and Shah 2015), INDEPENDENT SET (Brendel, Amer, and Todorovic 2011), MULTIGRAPH MATCHING (Hu et al 2020) and INTEGER QUADRATIC PROGRAMMING (Henschel et al 2018). In contrast to the two models considered here, the above examples often offer only limited options for integrating long-range interactions (Horňáková et al 2020).…”

Section: Introductionmentioning

confidence: 99%

The Complexity of Object Association in Multiple Object Tracking

Ganian

Hamm

Ordyniak

2021

AAAI

View full text Add to dashboard Cite

Object association, i.e., the identification of which observations correspond to the same object, is a central task for the area of multiple object tracking. Two prominent models capturing this task have been introduced in the literature: the Lifted Multicut model and the more recent Lifted Paths model. Here, we carry out a detailed complexity-theoretic study of the problems arising from these two models that is aimed at complementing previous empirical work on object association. We obtain a comprehensive complexity map for both models that takes into account natural restrictions to instances such as possible bounds on the number of frames, number of tracked objects and branching degree, as well as less explicit structural restrictions such as having bounded treewidth. Our results include new fixed-parameter and XP algorithms for the problems as well as hardness proofs which altogether indicate that the Lifted Paths problem exhibits a more favorable complexity behavior than Lifted Multicut.

show abstract

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

Cited by 11 publications

References 58 publications

LMGP: Lifted Multicut Meets Geometry Projections for Multi-Camera Multi-Object Tracking

LMGP: Lifted Multicut Meets Geometry Projections for Multi-Camera Multi-Object Tracking

Learnable Graph Matching: Incorporating Graph Partitioning with Deep Feature Learning for Multiple Object Tracking

The Complexity of Object Association in Multiple Object Tracking

Contact Info

Product

Resources

About