UvA-DARE is a service provided by the library of the University of Amsterdam (https://dare.uva.nl)
UvA-DARE (Digital Academic Repository)
Semi-interactive construction of 3D event logs for scene investigation
Dang, T.K.
Publication date 2013
Link to publication
Citation for published version (APA):
Dang, T. K. (2013). Semi-interactive construction of 3D event logs for scene investigation.
General rights
It is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s), other than for strictly personal, individual use, unless the work is under an open content license (like Creative Commons).
Disclaimer/Complaints regulations
If you believe that digital publication of certain material infringes any of your rights or (privacy) interests, please let the Library know, stating your reasons. In case of a legitimate complaint, the Library will make the material inaccessible and/or remove it from the website. Please Ask the Library: https://uba.uva.nl/en/contact, or a letter to: Library of the University of Amsterdam, Secretariat, Singel 425, 1012 WP Amsterdam, The Netherlands. You will be contacted as soon as possible.
Contents
1 Introduction 1
1.1 Motivation . . . 1
1.2 Problem Statement . . . 3
1.3 Organization . . . 4
2 A Review of 3D Reconstruction: Towards Scene Investigation Using Handheld Cameras 7 2.1 Overview of 3D Reconstruction From Video Sequences . . . 8
2.2 Lens Distortion Correction . . . 9
2.3 Feature Processing . . . 10
2.3.1 Interest Points . . . 11
2.3.2 Lines . . . 14
2.3.3 Initial Matching Strategy . . . 15
2.3.4 Summary and Conclusion . . . 16
2.4 Structure and Motion Recovery . . . 16
2.4.1 Multiple View Geometry and Stratification of 3D Geometry . . . 17
2.4.2 Projective Structure and Motion . . . 20
2.4.3 Metric Structure and Motion . . . 21
2.4.4 Degeneracy . . . 22
2.4.5 Reconstruction from Videos . . . 23
2.4.6 Summary and Conclusion . . . 24
2.5 Model Creation . . . 25
2.5.1 Rectification . . . 25
2.5.2 Stereo Mapping . . . 26
2.5.3 Mesh building and Texture mapping . . . 28
2.5.4 Discussion . . . 29
2.6 Conclusion and Discussion . . . 29
ii Contents 3 A Theoretical Analysis of The Perspective Error in Blob Detectors 31
3.1 Introduction . . . 32
3.2 Background on Blob Detectors . . . 32
3.2.1 The blob detector family . . . 33
3.2.2 Characteristics . . . 33
3.3 Perspective Drift . . . 34
3.3.1 Notation . . . 34
3.3.2 Perspective drift for one camera . . . 34
3.3.3 Relative perspective drift for two cameras . . . 37
3.3.4 Effects on 3D reconstruction . . . 38
3.4 Experiments . . . 39
3.4.1 Experimental setup . . . 39
3.4.2 Projected relative perspective drift . . . 40
3.4.3 Manifestations in reconstruction results . . . 42
3.4.4 Discussion . . . 45
3.5 Conclusion . . . 45
4 3D Modeling of Indoor Scenes Using Handheld Cameras 47 4.1 Introduction . . . 48
4.2 Background . . . 49
4.2.1 The 3D modeling process . . . 49
4.2.2 Degeneracy . . . 49
4.3 Modeling Framework for Indoor Scenes . . . 52
4.3.1 Frame Filter . . . 52
4.3.2 Frame Segmentation . . . 53
4.3.3 Structure recovery . . . 54
4.4 Results . . . 55
4.4.1 Evaluation of frame filtering . . . 55
4.4.2 Evaluation of frame segmentation . . . 58
4.4.3 Final result . . . 58
4.5 Conclusion . . . 59
5 A Semi-interactive Panorama Based 3D Reconstruction Framework for Indoor Scenes 61 5.1 Introduction . . . 62
5.2 Related Work . . . 63
5.2.1 Reconstruction from panoramas . . . 63
5.2.2 Interaction in reconstruction . . . 64
5.3 Framework Overview . . . 64
5.4 Building a Walls-And-Floor Model . . . 66
5.4.1 Smart corner picking . . . 67
5.4.2 Rectifying panoramas . . . 67
5.4.3 Estimating the floor-plan . . . 68
5.4.4 Reconstructability analysis . . . 70
Contents iii
5.5 Adding Details using Perspective Extrusion . . . 73
5.6 Results . . . 75
5.6.1 Datasets . . . 75
5.6.2 Accuracy . . . 75
5.6.3 Efficiency and Completeness . . . 77
5.7 Conclusion . . . 78
6 Building 3D Event Logs for Video Investigation 81 6.1 Introduction . . . 82
6.2 Related work . . . 85
6.3 Analyzing Investigation Logs . . . 86
6.3.1 Investigation events . . . 86
6.3.2 Segmentation using Structure-Motion Features . . . 87
6.4 Mapping Investigation Events to a 3D Model . . . 90
6.4.1 Automatic mapping of events . . . 91
6.4.2 Interactive mapping of events . . . 91
6.5 Evaluation . . . 92
6.5.1 Dataset . . . 93
6.5.2 Analyzing investigation logs . . . 94
6.5.3 Mapping events to 3D model . . . 96
6.6 Navigating Investigation Logs . . . 97
6.7 Conclusion . . . 98
7 Summary and Conclusion 101 7.1 Summary . . . 101
7.2 Conclusion and Future Work . . . 103
