Taming the Fire hose (Using visualization to explore data)

Preview:

DESCRIPTION

Taming the Fire hose (Using visualization to explore data). Penny Rheingans University of Maryland Baltimore County. Visualization Tasks. See values extrema anomalies boundaries/thresholds distribution / structure / pattern See multiple variables relationships See flow/change - PowerPoint PPT Presentation

Citation preview

Taming the Fire hoseTaming the Fire hose

(Using visualization to explore data)(Using visualization to explore data)

Penny Rheingans

University of Maryland Baltimore County

Visualization TasksVisualization TasksSee values

◦extrema◦anomalies◦boundaries/thresholds◦distribution / structure / pattern

See multiple variables◦relationships

See flow/changeUnderstand process

The Visualization ProcessThe Visualization Process

Refine

MappingControl

Volume VisualizationVolume Visualization Isosurface

◦ Pick important isolevel and render threshold surface

Direct volume rendering◦ Accumulate contributions of

voxels◦ Realistic rendering

Use physics-based illumination, accumulation, shadowing to enhance perception of data

◦ Transfer function design

Arbitrary mappings from scalar value to opacity

Volume RenderingVolume RenderingThreshold surfaces

Direct volume rendering

Photograph vs. IllustrationPhotograph vs. Illustration

Alice Tangerini, in Hodges89, pg 191

Illustration vs. Volume Illustration vs. Volume RenderingRendering

Volume IllustrationVolume IllustrationJoint work with David Ebert, Purdue

University, and various students (Chris Morris, Aidong Lu, Alark Joshi)

Approach◦Apply illustration techniques to volume

models to improve comprehensibility◦Provide toolbox of enhancement techniques

Issues◦What to show?◦How to show it?◦How to implement it?

Silhouette EnhancementSilhouette EnhancementSilhouette volumes are regions where

◦ there is a feature◦ feature normal is orthogonal to view

directionActions

◦ increase opacity◦ add outline

Implementation

ksc scales opacity of unenhanced features

kss controls size of maximum enhancement

kse controls rate of enhancement

Illuminated Gas Boundary and Illuminated Gas Boundary and Silhouette Silhouette

EnhancementEnhancement

Illuminated Gas Silhouette and Illuminated Gas Silhouette and Boundary Boundary EnhancementEnhancement

Understanding Understanding Change and Change and ProcessProcess

Flow IllustrationFlow IllustrationResearch done primarily by Alark JoshiAdding a dimension

◦ Volume: What’s inside this surface?◦ Temporal: How did it get this way?

Questions◦ Where was this feature before?◦ What is pattern of motion of features?◦ How has this feature changed?

Inspiration: Flow PathsInspiration: Flow Paths

McCloud

Application: Flow PathsApplication: Flow Paths

Inspiration: Strobe Inspiration: Strobe SilhouettesSilhouettes

McCloud

Application: Application: Strobe Strobe SilhouettesSilhouettes

A Very Short History of 3D A Very Short History of 3D VisualizationVisualization

Where?

What?

When?

Why?

WhatWhat’’s a Model?s a Model?

A system of postulates, data, and inferences presented as a mathematical description of an entity or state of affairs -- (12) Webster’s Ninth Collegiate Dictionary

Data vs Model◦ Discrete : continuous◦ Deterministic : probabilistic◦ Examples : explanations

Thinking with ModelsThinking with Models

Modelco

nstr

uctconcept

generateVisual repvisualize

Visual repvisualizeInput

data

sam

ple

Outcome data

visualizeVisual rep

val i

dat e

vali

date

vali

date

refine

Some ChallengesSome Challenges Understanding models

◦ Sampling and interrogating◦ High-dimensional structure of model◦ Model sensitivity and variability

Comparing models◦ Validating outcome against reality◦ Validating model against from reality◦ Examining families of models

Constructing models◦ Selecting parameters◦ Incremental construction◦ Abstract specification

Questions?Questions?