Wissenschaftliche Grundlagen 3D Stadtmodelle

1
A Concept ofEffective Landmark Depictionin
Geovirtual 3D Environments by View-Dependent
Deformation Matthias Trapp Hasso-Plattner-In
stitut Computer Graphics Systems Group Prof. Dr.
Jürgen Döllner University Potsdam www.hpi.uni-pot
sdam.de/3d www.3dgi.de
2

• Motivation
• Conceptual Overview
• Tagging Geometric Preprocessing
• Deformation Model
• Rendering Technique
• Conclusions

3
1. Motivation

• Tourist Map (St. Petersburg)

Source www.escapetravel.spb.ru
4
1. Motivation Previous Related Work

• Guidelines for landmark depictions (Vinson 1999)
• LM should be identifiable and distinguishable
• Ascertained instead of abstract representations
• LM should be visible in all map scales
• Previous approaches
• Use photos instead of abstract depiction
• Iconification / application of symbols
• Scaled depiction of landmarks in 2D

Lee 2001
Baudisch 2003
Elias 2006
5
1. Motivation Visibility Problem

• Restricted visibility of landmark objects in
  interactive systems
• Depicted landmarks are too small/far away
• Occluded by unimportant buildings

6
1. Motivation Solution for Interactive Systems
7

• Motivation
• Conceptual Overview
• Tagging Geometric Preprocessing
• Deformation Model
• Rendering Technique
• Conclusions

8
2. Conceptual Overview

• Integration into visualization pipeline (Ware
  2000)
• 3-Phase Process
• Tagging augment city model with landmark
  information
• Preprocessing transform city model to render
  geometry
• Rendering rendering of processed scene geometry

9

• Motivation
• Conceptual Overview
• Tagging Geometric Preprocessing
• Deformation Model
• Rendering Technique
• Conclusions

10
3. Tagging Geometric Preprocessing

• Tagging a City Model
• Associate weight w(ci) to a city model object ci
• Defining importance function
• Partition of C into landmarks and non-landmark
  objects
• Augmentation can be automated via services

11
3. Tagging Geometric Preprocessing

• Input
• Tagged city model
• Output
• Landmark deformation data
• City model geometry

12

• Motivation
• Conceptual Overview
• Tagging Geometric Preprocessing
• Deformation Model
• Rendering Technique
• Conclusions

13
5. Deformation Model

• Tasks
• Landmark scaling
• Displacement of surrounding buildings
• Mutual landmark displacement
• Input

14
5. Deformation Model Landmark Scaling

• Quadratic scaling function
• Scaling depend on distance object-to-camera d
• Property defined by scale interval
• Interval derived from weight w(ci)

Projected size of a landmark
Scaling function
15
5. Deformation Model Displacement

• Displacement of environment
• Prototypic radial displacement
• Translation of surrounding buildings
• Inverse linear scaling
• Mutual landmark displacement
• Naive spring model (Bobrich 1996)
• Topology is not preserved
• Problems with frame-to-frame coherence
• ? Limiting factor of this landmark visualization

16

• Motivation
• Conceptual Overview
• Tagging Geometric Preprocessing
• Deformation Model
• Rendering Technique
• Conclusions

17
6. Rendering Technique

• Rendering Requirements
• Programmable graphics hardware (Vertex Shader)
• Scene graph-based rendering system
• Multi-Pass evaluation
• Pre-Traversal Pass
• Collect all landmark attribute nodes
• Perform view-frustum culling to nodes
• Application of deformation model
• Rendering Pass Apply deformation results
• Shape-preserving (uniform) scaling
• Per-Vertex displacement

18

• Motivation
• Conceptual Overview
• Tagging Geometric Preprocessing
• Deformation Model
• Rendering Technique
• Conclusions

19
6. Conclusions

• Characteristics
• Interactive, view-dependent landmark depictions
• Fully accelerated by modern 3D graphics hardware
• Uses vertex deformation and displacement
• Open Problems Future Work
• Improvement coherency of deformation model
• Research alternative deformation variants
• Non-uniform deformation

20
Contact

• Matthias Trappmatthias.trapp_at_hpi.uni-potsdam.de
• Computer Graphics Systems GroupProf. Dr. Jürgen
  Döllnerwww.hpi.uni-potsdam.de/3d
• Researchgroup 3D-Geoinformationwww.3dgi.de

21
References

• Elias, Paelke, Kuhnt, Kartographische
  Visualisierung von Landmarken, Aktuelle
  Entwicklungen in Geoinformation und
  Visualisierung, GEOVIS 2006, 5./6. April 2006,
  Potsdam, Kartographische Schriften Band 10, 2006.
• Vinson, Design Guidelines for Landmarks to
  Support Navigation in Virtual Environments,
  Proceedings of CHI 99, ACM, 1999.
• C. Ware, Information Visualization Perception
  for Design, Morgan Kaufmann, 2000
• J.Bobrich, Ein neuer Ansatz zur kartographischen
  Verdrängung auf der Grundlage eines mechanischen
  Federmodells, Ph.D. Thesis, Vol. C 455, Deutsche
  Geodätische Kommission, München, 1996
• P. Baudisch and Ruth Rosenholtz, Halo a
  Technique for Visualizing Off-Screen Locations,
  Proceedings of the ACM SIGCHI conference on Human
  factors in computing systems, pp. 481-488, 2003
• Y.C. Lee, A. Kwong, L. Pun and A. Mack,
  Multi-Media Map for Visual Navigation, Journal
  of Geospatial Engineering, Hong Kong, 2001