Jump to content
Higher-Order Multiple Object Tracking / Titelei/Inhaltsverzeichnis
Higher-Order Multiple Object Tracking / Titelei/Inhaltsverzeichnis
Contents
Chapter
Expand
|
Collapse
Page
Titelei/Inhaltsverzeichnis
Details
1–32
1 Introduction
1–32
Details
1.1 Applications
Details
1.2 The Multiple Object Tracking Problem
Details
1.2.1 Video-based Multiple Object Tracking (MOT)
Details
1.2.2 Tracking-by-detection
Details
1.3 Challenges of Multiple Object Tracking
Details
1.3.1 Errors caused by the object detector
Details
1.3.2 Challenges in discriminative features
Details
1.3.3 Combinatorial challenges
Details
1.4 Related Work
Details
1.5 Contributions
Details
1.6 List of Publications
Details
1.7 Outline
Details
33–79
2 Fundamentals
33–79
Details
2.1 Sets, Maps, and Matrices
Details
2.2 Probability Theory
Details
2.3 Graph Theory
Details
2.3.1 Important graph classes
Details
2.3.2 Computations on graphs
Details
2.4 Machine Learning
Details
2.4.1 Supervised learning
Details
2.4.2 Logistic regression
Details
2.4.3 Neural networks
Details
2.5 Computational Complexity Theory
Details
2.6 Optimization Theory
Details
2.6.1 Linear programming
Details
2.6.2 Binary linear programming
Details
2.6.3 Quadratic programming
Details
2.6.4 Binary quadratic programming
Details
2.6.5 Non-linear optimization
Details
2.7 Multi-Object Tracking
Details
2.7.1 Object detectors
Details
2.7.2 Appearance features
Details
2.7.3 Datasets
Details
2.7.4 MOT metrics
Details
80–129
3 HO-MOT with Signal Fusion
80–129
Details
3.1 Introduction
Details
3.2 Signal Fusion as Weighted Graph Labeling Problem
Details
3.2.1 Related work
Details
3.2.2 Data association model for signal fusion
Details
3.3 Frank-Wolfe Optimizer for Weighted Graph Labeling Problems
Details
3.3.1 Related work
Details
3.3.2 Frank-Wolfe Optimizer for Binary Solutions
Details
3.4 Multiple People Tracking by Fusing Head and People Detections
Details
3.4.1 Data association model
Details
3.4.2 Experimental results
Details
3.5 Simultaneous Identification and Tracking of Multiple People using Video and IMUs
Details
3.5.1 Related work
Details
3.5.2 Method
Details
3.5.3 Evaluation
Details
3.6 Conclusion
Details
130–164
4 Lifted Disjoint Paths
130–164
Details
4.1 Introduction
Details
4.2 Related Work
Details
4.3 Problem Formulation
Details
4.4 Constraints
Details
4.5 Separation
Details
4.6 Complexity
Details
4.7 Experiments
Details
4.7.1 Graph construction
Details
4.7.2 Pre-processing and post-processing
Details
4.7.3 Cost learning
Details
4.7.4 Implementation details on the lifted disjoint paths solver
Details
4.7.5 Experiment setup
Details
4.7.6 Benefit of long-range edges
Details
4.7.7 Ablation study on post-processing methods
Details
4.7.8 Accuracy of the fusion network
Details
4.7.9 Qualitative evaluations
Details
4.7.10 Benchmark evaluations
Details
4.8 Conclusion
Details
165–171
5 Conclusions
165–171
Details
172–196
Bibliography
172–196
Details
Durchsuchen Sie das Werk
Geben Sie ein Keyword in die Suchleiste ein
Partial access
Higher-Order Multiple Object Tracking
Titelei/Inhaltsverzeichnis
Autoren
Roberto D. Henschel
DOI
doi.org/10.51202/9783186875105-I
ISBN print: 978-3-18-387510-8
ISBN online: 978-3-18-687510-5
Chapter Preview
Share
Download PDF
Download citation
RIS
BibTeX
Copy DOI link
doi.org/10.51202/9783186875105-I
Share by email
Video schließen
Share by email VDI Verlag eLibrary
Recipient*
Sender*
Message*
Your name
Send message
This site is protected by reCAPTCHA and the Google
Privacy Policy
and
Terms of Service
apply.