GIAnT: Visualizing Group Interaction at Large Display Walls Ulrich - - PowerPoint PPT Presentation

Ulrich von Zadow, Raimund Dachselt

GIAnT:

Visualizing Group Interaction at Large Display Walls

CHI, May 2017

ACM CHI 2017 – IML Dresden – GIAnT: Visualizing Group Interaction at Large Wall Displays

2

Display walls: § Very large surfaces § Interactive: Touch, tangibles, pen, distal pointing,... § Very high resolution § Users move around § Multiple users § Public/semi-public Background: Display Walls

CityWall [Morrison et al., 2008] Wild Room [Beaudouin-Lafon et al., 2008] GraSp [Kister et al., 2017] NiCE Room [Haller et al., 2008]

ACM CHI 2017 – IML Dresden – GIAnT: Visualizing Group Interaction at Large Wall Displays

3

Application Cases

[Rajabiyazdi et al., 2015]

Locomotion [Ball et al., 2008] Collaboration

[Jakobsen and Hornbæk, 2014]

Awareness

[von Zadow et al., 2016]

Background: Research in Display Wall Interaction

ACM CHI 2017 – IML Dresden – GIAnT: Visualizing Group Interaction at Large Wall Displays

4

Research Question:

What visualizations would allow us to understand multi-user interaction at large display walls more effectively? Challenges: § Support time periods: Provide visualizations that show user behavior

• ver time periods

§ Provide dimensional mapping: Display high-dimensional data (e.g., 3D position + time) on a 2D display § Support relevant research topics, e.g., collaboration, locomotion, awareness

ACM CHI 2017 – IML Dresden – GIAnT: Visualizing Group Interaction at Large Wall Displays

5

Video Coding Tools VCode and VData

[Hagedorn et al., 2008]

Tabletop Visualization Toolkits VisTaCo

[Tang et al., 2010]

Related Work: Interaction Analysis Tools

ACM CHI 2017 – IML Dresden – GIAnT: Visualizing Group Interaction at Large Wall Displays

6

Group Interaction Analysis Toolkit GIAnT git repository: https://github.com/imldresden/GIAnT

ACM CHI 2017 – IML Dresden – GIAnT: Visualizing Group Interaction at Large Wall Displays

7

Time Slider

ACM CHI 2017 – IML Dresden – GIAnT: Visualizing Group Interaction at Large Wall Displays

8

Video

ACM CHI 2017 – IML Dresden – GIAnT: Visualizing Group Interaction at Large Wall Displays

9

Timeline View

ACM CHI 2017 – IML Dresden – GIAnT: Visualizing Group Interaction at Large Wall Displays

10

Wall View

ACM CHI 2017 – IML Dresden – GIAnT: Visualizing Group Interaction at Large Wall Displays

11

Floor View

ACM CHI 2017 – IML Dresden – GIAnT: Visualizing Group Interaction at Large Wall Displays

12

Statistics

ACM CHI 2017 – IML Dresden – GIAnT: Visualizing Group Interaction at Large Wall Displays

13

Colors § L*C*h Color Space § Different hues § Equal perceptual brightness Time Position across wall Visualizations: Timeline View Distance from wall White dots: Touches

ACM CHI 2017 – IML Dresden – GIAnT: Visualizing Group Interaction at Large Wall Displays

14

Visualizations: Alternative Timeline Time Distance from wall White dots: Touches

ACM CHI 2017 – IML Dresden – GIAnT: Visualizing Group Interaction at Large Wall Displays

15

Colors § Additive blending § Regions with

• verlapping gaze

are brighter Visualizations: Wall View § Color-coded heat map of gaze points § Scatterplot of touches

ACM CHI 2017 – IML Dresden – GIAnT: Visualizing Group Interaction at Large Wall Displays

16

§ Heat map of user positions § Current positions as dots, viewpoints as lines Visualizations: Floor View

ACM CHI 2017 – IML Dresden – GIAnT: Visualizing Group Interaction at Large Wall Displays

17

Live Demo

ACM CHI 2017 – IML Dresden – GIAnT: Visualizing Group Interaction at Large Wall Displays

18

ACM CHI 2017 – IML Dresden – GIAnT: Visualizing Group Interaction at Large Wall Displays

19

GIAnT System Design

Data Sources Derived data Visualizations

Timeline Wall Floor Statistics User Pos Gaze Dir Touch Video

ACM CHI 2017 – IML Dresden – GIAnT: Visualizing Group Interaction at Large Wall Displays

20

Abstract data sources Concrete data sources GIAnT System Design Data Sources

User Pos Gaze Dir Touch Video

Marker-based Tracking User coding using tangible markers

ACM CHI 2017 – IML Dresden – GIAnT: Visualizing Group Interaction at Large Wall Displays

21

Technologies § Python/libavg § SQLite database § C++ visualizations for speed § Preprocessor script GIAnT System Design Open Source § https://github.com/imldresden/GIAnT § GPL license § Any issues? Contact us!

ACM CHI 2017 – IML Dresden – GIAnT: Visualizing Group Interaction at Large Wall Displays

22

Set of Visualizations for Wall Interaction Analysis

§ Shows user interaction over configurable time periods § Intelligent dimensional mapping § Supports relevant research topics

Group Interaction Analysis Toolkit GIAnT

§ Open Source § Expandable to new visualizations, data sources, derived data Conclusion: Contributions

Th Thank ank yo you!

Contact: Ulrich von Zadow ulrich.zadow@tu-dresden.de

GIAnT GIAnT:

Visualizing Group Interaction at Large Display Walls

https://www.imld.de/giant https://github.com/imldresden/GIAnT

ACM CHI 2017 – IML Dresden – GIAnT: Visualizing Group Interaction at Large Wall Displays

24

References

ACM CHI 2017 – IML Dresden – GIAnT: Visualizing Group Interaction at Large Wall Displays

25

• Shows derived data
• Parallel coordinates emphasize correlations
• Candidate for expansion

Visualizations: Statistics View

ACM CHI 2017 – IML Dresden – GIAnT: Visualizing Group Interaction at Large Wall Displays

26

Extrafolien:

• Research topics
• ...

...

ACM CHI 2017 – IML Dresden – GIAnT: Visualizing Group Interaction at Large Wall Displays

27

• Colors determined in L*C*h space.
• Users differentiated by hue, equal

perceptual brightness and distance.

• Lightness available to visualize other

variables.

• Additive blending results in clearly visible
• verlaps (e.g. heatmaps)

Visualizations: Colors