Design by Dragging: An Interface for Creative Forward and Inverse Design with Simulation Ensembles

Abstract

We present an interface for exploring large design spaces as encountered in simulation-based engineering, design of visual effects, and other tasks that require tuning parameters of computationally-intensive simulations and visually evaluating results. The goal is to enable a style of design with simulations that feels as-direct-as-possible so users can concentrate on creative design tasks. The approach integrates forward design via direct manipulation of simulation inputs (e.g., geometric properties, applied forces) in the same visual space with inverse design via “tugging” and reshaping simulation outputs (e.g., scalar fields from finite element analysis (FEA) or computational fluid dynamics (CFD)). The interface includes algorithms for interpreting the intent of users’ drag operations relative to parameterized models, morphing arbitrary scalar fields output from FEA and CFD simulations, and in-place interactive ensemble visualization. The inverse design strategy can be extended to use multi-touch input in combination with an as-rigid-as-possible shape manipulation to support rich visual queries. The potential of this new design approach is confirmed via two applications: medical device engineering of a vacuum-assisted biopsy device and visual effects design using a physically based flame simulation.

Visualization System Requirements for Data Processing Pipeline Design and Optimization

“…Each colored cell indicates a requirement that is important for a given case study/paper, and an 'x' indicates that the requirement remains unmet by users' current tools. [25] req req req req req Chemical process models [26] x req req x Economic modelling [27] req req xx req req x req Aircraft engine design [28] req req x req req Phylogenetic trees [29,30] x req x x req req Molecular evolution [31,32] x x req x req x * Raidou et al [75] iCoCooN req req req * Ruppert et al [46] --x req x Luboshik et al [38] --x req req req Bruckner et al [10] --x req x req * Beham et al [43] Cupid req req xx Konyha et al [39] --x req x Pretorius et al [12] Paramorama req x req Afzal et al [36] RVF x req req x Bergner et al [13] ParaGlide xx req Torsney-Weir et al [33] Tuner x req x x * Padua et al [11] --x req x * Bögl et al [42] TiMoVA req req req x x req Spence et al [73] --req req Berger et al [9] --req x xx Piringer et al [40] HyperMoVal xx req req req Matkovic et al [17] --req x Coffey et al [34] --x x req Matkovic et al [50] --x Potter et al [49] Ensemble-Vis x req Booshehrian et al [19] Vismon req x req x Brecheisen et al [52] --x req x Unger et al [54] --req req req Amirkhanov et al [74] --x Marks et al [48] Design Galleries req x Waser et al [18] World Lines req x x * Martins et al [35] --x req x * Wu [53] --xx x Guo et al [41] --req…”

“…First, unstructured data is inherently more difficult to analyze than data that are structured into the rows and columns of a conventional database. Examples include free text, images from a multitude of patients [33], multimedia, and the designs of 3D models [34]. Second, as the number of variables increases (e.g., [35]), it is more challenging to deal with inputs.…”

“…Alternatively, the pipeline may contain a number of 'what if' branches [42]. Sometimes individual parameters are easy for users to understand (e.g., the number of clusters for k-means), but other parameters are abstractions of physical systems [34] or the parameters are difficult to relate to output consequences [11], [46]. Sometimes, users tend to treat the pipeline as a 'black box' and accept the default settings [47].…”

<italic>Cupid</italic>: Cluster-Based Exploration of Geometry Generators with Parallel Coordinates and Radial Trees

“…The user can deform a template model and the corresponding object is shown. Coffey et al [9] extend this technique to explore simulations or visual effects. By dragging parts of an object, the user can explore different simulations.…”

“…By dragging parts of an object, the user can explore different simulations. Smith et al [38] and Coffey et al [9] also integrate abstract data into the visualization. Smith et al add attributes to the generated cars such as weight and sportiness.…”