130 likes | 152 Views
Perceptual Integration in Visualization Environments. Project Proposal Nicholas F. Polys, Ph.D. Virginia Tech Research Computing. Industry Trends. Emerging standards are casting 4D data as a first-class citizen of the information enterprise.
E N D
Perceptual Integration in Visualization Environments Project Proposal Nicholas F. Polys, Ph.D. Virginia Tech Research Computing
Industry Trends • Emerging standards are casting 4D data as a first-class citizen of the information enterprise. • 3D models, simulation results, and virtual environments can be annotated and linked to other information such as cost/tolerance databases and semantic repositories. • For example Web3D Consortium’s active development of the CAD profile including geometry and metadata (www.web3d.org), ISO, W3C
Recent Research – Display Venues • VT Computer Science, Center for HCI show high-res and immersive display venues CAN improve task performance: • Analyze 22x more data in only 3x more time while maintaining accuracy • Reduce virtual navigation actions by 75% • Reduce frustration by 50% • Short initial learning time
New Opportunities – Display Techniques • Spatial, Abstract, and Temporal data can be combined, delivered and presented in an ‘integrated information space’ • Attributes and annotations plus objects and groups can be rendered with a variety of (in)consistent perceptual cues
The Challenge • The real digital divide is the last ten feet between the interface and the mind Making Sense Perception Interpretation Proposed design will cost too much in long term maintenance color, shading, lines characters, squares, spatial organization (Working Memory) Excel worksheet, a part is selected, formula is displayed at top
Recent Research • “Are this annotation and this referent related?” • Information Theoretic approach to information layout shows complex interactions of perceptual cues depending on task, data target, and display context
A Proposal • Prior work in Visual Computing has not been applied to engineering design review applications • Depth and Gestalt cues interact differently with stereo, surround, or head-tracked displays • Two Phase approach: • Apply layout testbed for design-review data across desktop, wall, and CAVE display • Run user studies measuring objective (e.g. accuracy, time) and subjective metrics (e.g. cognitive load, satisfaction) during task performance
Phase 1 of 2 Software testbed • Evaluate design review requirements • Adapt IRVE tools to design review application • CAD models • Data types & graphs • Port and test across display venues
Phase 2 of 2 User Studies • What effects do different combinations of perceptual cues have on feature binding in working memory? • How can these cue combinations be optimized per task? • How can these cue combinations be optimized per display (e.g. size, stereo, head-tracking)?
Direct Impacts • From our study data, we will construct a cognitive model of the interaction of perceptual cues across tasks and displays • The model may drive the interface to adaptively render heterogeneous information depending on the task and display • This will improve the efficiency and accuracy of design review interfaces across: • Desktop displays (mono) • Wall displays (mono, stereo, head tracking) • CAVE (mono, stereo, head tracking)
Broader Impacts • Integrated visualization capabilities are necessary for users to gain a full understanding of complex relationships in their heterogeneous data • Application designers must take account of how humans build their cognitive models and what perceptual predispositions and biases are in play • With such knowledge, designers can take steps to minimize or leverage their effect and create advantageous research, design, and decision-support applications
Thank you! Contact: Nicholas F. Polys Ph.D. VT Research Computing npolys@vt.edu