1 / 55

Visualization (part II)

Visualization (part II). Nicholas F. Polys, Ph.D. VT Research Computing. Norman’s Gulfs. Don Norman, 1986. Making Sense of an Information Display. Interpretation. Excel worksheet, a cell is selected, formula is displayed at top. Making Sense. Perception. Income worksheet,

michelleerickson
Download Presentation

Visualization (part II)

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript

  1. Visualization (part II) Nicholas F. Polys, Ph.D. VT Research Computing

  2. Norman’s Gulfs Don Norman, 1986

  3. Making Sense of an Information Display Interpretation Excel worksheet, a cell is selected, formula is displayed at top Making Sense Perception Income worksheet, Total tax income is being calculated, the wrong multipler is being used color, shading, lines characters, squares, spatial organization Last month’s budget... ?

  4. Making Sense • Last step in crossing the Gulf of Evaluation • Information has been perceived and interpreted • Users must “make sense” of information by relating it to their tasks, goals, and interests • Designers must support people’s abilities to detect patterns and relationships • Consistent use of shape, size, color, position • Information models (e.g., hierarchies) organize data • Dynamic displays cue users to structure

  5. Important Considerations • Understanding the domain • Understanding the Research Question • Understanding the purpose of the Vis • User and reader tasks • Including descriptive captions and legends

  6. Which network is easier to understand? See: http://www.graphviz.org/

  7. Context Required

  8. Tools • There are many (e.g. IDL, Amira, TecPlot, VMD) • This afternoon we’ll explore some powerful, general-purpose, open-source tools from National Labs - Paraview www.paraview.org - VisiT https://wci.llnl.gov/codes/visit/home.html

  9. Proprietary vs. Opensource Who owns the data? Who owns the tools to access that data? How are bugs/new features accomplished? How much does it cost? Games, Google, Second Life vs. open standards

  10. Trends in Visualization • GPUs power increases (for graphics and other floating point computations such as CUDA) • Integrating analysis, AI, and stats into visualization tools • Data goes large

  11. The Data Problem • HPC runs generate massive amounts of output but we want to represent and explore it in an interactive way • Its heavy to move • Some approaches: • Write out subsamples for vis during simulation • Reduce or subsample existing output • Preprocess output to Vis • Sneaker Net vs. Remote Vis

  12. Visual Analysis Overview User Raw data Tables Visual structures Views Data transforms | Visual attribute | View transforms | Rendering assignment Figure 2.1: Processing in a typical visualization pipeline (from Card et al, 1999)

  13. Reduce Data • Write out subsamples for vis during simulation • Reduce or subsample existing output

  14. Visual Analysis Overview User Raw data Tables Visual structures Views Data transforms | Visual attribute | View transforms | Rendering assignment Figure 2.1: Processing in a typical visualization pipeline (from Card et al, 1999)

  15. Preprocess Preprocess output to Vis + -

  16. PathSim • PathSim is a computer model and simulation engine designed for Systems Biology investigators and Virologists to study the dynamics of an immune system under various infection conditions in silico • Agent-based Simulations on anatomical geometry with biological agent interactions, set from initial physiological conditions Polys, Nicholas F., Bowman, D., North, C., Laubenbacher, R., Duca, K., (2004). PathSim Visualizer: An Information-Rich Virtual Environment for Systems Biology. Web3D Symposium, Monterey, CA, ACM Press.

  17. PathSim Application • Managing and Visualizing large, multi-scale agent-based simulation results through the web • Extensive use of annotation concept for views on: data, metadata, networked multimedia • Scaling user-space controls across scales • Generating insight into system dynamics for diagnosis, and ‘what if?’ to model interventions

  18. PathSim Data • Massive: upwards of 7 million agents whose state and location can be measured every 6 simulation minutes!

  19. PathSim Visualizer Data Store Viz Processors Spatial data [XML: sim_enviro.dtd organ.dtd tissue.dtd] View V V View View V Abstract data Temporal data [agent population timeseries] Integrated User View Overview + Detail

  20. Application Example:PathSim Anatomy A micro-scale, annotated view of a tonsil tissue mesh Tissue Organ Part, Organ A micro-scale VRML view of the unit section tissue mesh translated from XML

  21. Application Example:Metadata and Annotation Layout PathSim Macro & Micro views of Agent-based simulation results

  22. Visual Analysis Overview User Raw data Tables Visual structures Views Data transforms | Visual attribute | View transforms | Rendering assignment Figure 2.1: Processing in a typical visualization pipeline (from Card et al, 1999)

  23. Remote Visualization Data and rendering on some big remote resource + -

  24. Services & Servers • Integrated databases • Interoperable file formats • Referenced resources across the web • Visualization middleware services • Multi-user & persistent worlds

  25. Integrating Visualization into Proposals Improved visualization support is a recognized challenge: • NSF / NIH Report - 2006 • Visual Analytics Initiative - 2006 • Many other agencies are facing the same problem: making sense of large, heterogeneous data sets

  26. NSF / NIH Visualization Report A renewed funding priority for basic research- transformative technology and techniques http://tab.computer.org/vgtc/vrc/index.html • C. Johnson, R. Moorhead., T. Munzner, H. Pfister, P. Rheingans, and T. S. Yoo, (Eds.): (2006). NIH-NSF Visualization Research Challenges Report, IEEE Press).

  27. Visual Analytics R&D Agenda A renewed initiative in visualization, recasting the problem to interactive analysis tools for large, complex data sets http://nvac.pnl.gov/agenda.stm • Thomas, J. J., and Cook, Kristin A. (2006). A Visual Analytics Agenda. IEEE Computer Graphics & Applications, 10-13.

  28. NAE Grand Challenges • National Academy of Engineering notes that improving Virtual Reality is a grand challenge (worthy of pursuit)! http://www.engineeringchallenges.org/cms/8996/9140.aspx

  29. Educational Technology Horizon Report 2007– New Media Consortium • Proprietary and open technology exists! • Adoption timeline: • 1-2 years Virtual Worlds • 3-5 years Multi-User spaces http://www.nmc.org/horizon/

  30. VT CAVE Immersive visualization venue • 3 wall + floor stereo projected surround • Head & Input tracking • Not a cost center! (Free use for faculty and student projects) • www.cave.vt.edu

  31. Stereo Walls TORG (active) Andrews (passive) 3-4 additional in labs around campus

  32. VT GigaPixel Laboratory • ~200 Mpixels, scalable • Reconfigurable • Multiple display technologies • Diverse input devices • Link to AwareLab, VICON

  33. Analyst Workspace Options 9 tiled LCD panels Single CPU MS Windows Small cubicle < $5,000 24 tiled LCD panels Curved configuration Multi-CPU Large cubicle/office ~ $20,000

  34. Collaborative Workspace Options 18 Rear-projection blocks near-seamless Multi-CPU Large conference room ~ $150,000 50 tiled LCD panels Touch-sensitive Multi-CPU Medium conference room ~ $120,000

  35. Matching Tool and Researcher • Karen Duca, VBI • Effect of smoking on Flu immune system? • >3 million data values • Spotfire™, …

  36. Matching Tool and Researcher • Fisheries & Wildlife Science • Computational Fluid Dynamics

  37. Matching Tool and Researcher • Molecular Dynamics Simulation (NIH)

  38. Develop New Vis Tools • NuTech, Inc. Geospatial agent-based reasoning • Research tool • Parallel processing • Visualization front end

  39. Develop New Vis Tools • Server-side system (DARPA) • ICAM satellite flux modeling

  40. “Look Ma, no equations!”

  41. Develop New Vis Tools • mpiBLAST DNA matching visualization

  42. Engineer New Vis Products • Census Bureau • Emphasis on UE process, users/customers • Distributable Tool • Counties USA CD-ROM

  43. Engineer New Vis Products • PathSim (VBI- Duca, Laubenbacher, Tufts U) • Agent-based immune system simulation + visual analysis environment • Network-accessibility for timeseries data

  44. PathSim Web-based visualization & analytic services upwards of 7 million agents whose state and location can be measured every 6 simulation minutes!

  45. Basic Research • Bill Carstensen, Dept of Geography and CGIT • How do large high-resolution displays & visualizations impact geospatial analysts? Some key results…

  46. Faster User Task Performance Display size vs. User’s task performance time Users were ~8 times faster with 8 times more screen space p=0.01  Faster Larger display size →

  47. Key Results • ↑Display size → ↓Virtual navigation, ↑Physical navigation, ↓Performance time • Up to 10x task performance boost • Virtual Nav has more impact on performance • Analysts prefered Physical Nav (100% when zooming choice) • Differing search strategies/paths • Tethered analyst → reduced benefit

  48. Basic Research Carilion Biomedical – Karen Roberto, Center for Gerontology, UVA Medical • Web + CAVE virtual worlds as assessment tools for Mild Cognitive Impairment

  49. Key Results • Open standards enabled cross-platform testing • Immersive technologies can provide powerful presence for users • Safe and private assessment tool

More Related