Sundar Murugappan

Extended multitouch: recovering touch posture and differentiating users using a depth camera

Multitouch surfaces are becoming prevalent, but most existing technologies are only capable of detecting the users actual points of contact on the surface and not the identity, posture, and handedness of the user. In this paper, we define the concept of extended multitouch interaction as a richer input modality that includes all of this information. We further present a practical solution to achieve this on tabletop displays based on mounting a single commodity depth camera above a horizontal surface. This will enable us to not only detect when the surface is being touched, but also recover the users exact finger and hand posture, as well as distinguish between different users and their handedness. We validate our approach using two user studies, and deploy the technique in a scratchpad tool and in a pen + touch sketch tool.

FEAsy: A Sketch-Based Interface Integrating Structural Analysis in Early Design

The potential advantages of freehand sketches have been widely recognized and exploited in many fields especially in engineering design and analysis. This is mainly because the freehand sketches are an efficient and natural way for users to visually communicate ideas. However, due to a lack of fundamental techniques for understanding them, sketch-based interfaces have not yet evolved as the preferred computing platform over traditional menu-based tools. In this paper, we address the specific challenge of transforming informal and ambiguous freehand inputs to more formalized and structured representations. We present a domain-independent, multi-stroke, multi-primitive beautification method which detects and uses the spatial relationships implied in the sketches. Spatial relationships are represented as geometric constraints and satisfied by a geometric constraint solver. To demonstrate the utility of this technique and also to build a natural working environment for structural analysis in early design, we have developed FEAsy (acronym for F inite E lement A nalysis made easy ) as shown in Fig. 1. This tool allows the users to transform, simulate and analyze their finite element models quickly and easily through freehand sketching, just as they would draw on paper. Further, we have also developed simple, domain specific rules-based algorithms for recognizing the commonly used symbols and for understanding the different contexts in finite element modeling. Finally, we illustrate the proposed approach with a few examples.Copyright

Towards beautification of freehand sketches using suggestions

Beautification of freehand sketches is integral for building robust sketch understanding systems and sketch-based interfaces for CAD. Many of the current methods for beautification do not consider some important information implied in the sketches such as spatial relationships (geometric constraints) between primitives. In addition, as the freehand input is ambiguous in nature, correctly interpreting the visual scene the user has in mind is a difficult problem. To this extent, we present our ongoing work, a suggestive interface for constraint-driven beautification of freehand sketches which provides multiple interpretations of the freehand input, from which the user can choose the intended result. A preliminary user study has been conducted to evaluate the effectiveness of the proposed method.

Towards locally and globally shape-aware reverse 3D modeling

The process of re-creating CAD models from actual physical parts, formally known as digital shape reconstruction (DSR) is an integral part of product development, especially in re-design. While, the majority of current methods used in DSR are surface-based, our overarching goal is to obtain direct parameterization of 3D meshes, by avoiding the actual segmentation of the mesh into different surfaces. As a first step towards reverse modeling physical parts, we extract (1) locally prominent cross-sections (PCS) from triangular meshes, and (2) organize and cluster them into sweep components, which form the basic building blocks of the re-created CAD model. In this paper, we introduce two new algorithms derived from Locally Linear Embedding (LLE) (Roweis and Sauk, 2000 [3]) and Affinity Propagation (AP) (Frey and Dueck, 2007 [4]) for organizing and clustering PCS. The LLE algorithm analyzes the cross-sections (PCS) using their geometric properties to build a global manifold in an embedded space. The AP algorithm, then clusters the local cross sections by propagating affinities among them in the embedded space to form different sweep components. We demonstrate the robustness and efficiency of the algorithms through many examples including actual laser-scanned (point cloud) mechanical parts.

Virtual Repair: Geometric Reconstruction for Remanufacturing Gas Turbine Blades

The usage of Direct Metal Deposition (DMD) technology for repairing defective voids in high-value metallic components is time consuming, since traditional geometric reconstruction methods are not seamlessly connected to the DMD process. Here, we consider the development of a semi-automated geometric algorithm for “virtually” repairing defective voids that appear on gas turbine airfoils after extensive use. Our method produces an accurately reconstructed geometric model that can be used for generating DMD scanning paths, while ensuring both dimensional accuracy and topological consistency required by the airfoil design. This model is constructed by using the Sectional Gauss Map concept to generate a series of Prominent Cross Sections (PCS) along the longitudinal axis of a digitally acquired defective airfoil. The intrinsic geometry of the PCS lying in the non-defective region is then extrapolated across the defective region to fill in the voids. A boolean difference between the original defective model and the final reconstructed model yields a fully parameterized geometric representation of the repair volume. The test results of this method demonstrate the algorithm’s robustness and versatility over a wide range of airfoil defects.

Handy-Potter: Rapid Exploration of Rotationally Symmetric Shapes Through Natural Hand Motions

We present a paradigm for natural and exploratory shape modeling by introducing novel 3D interactions for creating, modifying and manipulating 3D shapes using arms and hands. Though current design tools provide complex modeling functionalities, they remain non-intuitive for novice users. Significant training is required to use these tools since they segregate 3D shapes into hierarchical 2D inputs, binding the user to stringent procedural steps and making modifications cumbersome. On the other hand, the use of CAD systems is typically involved during the final phases of design. This leaves a void in the early design phase wherein the creative exploration is critical. We present a shape creation paradigm as an exploration of creative imagination and externalization of shapes, particularly in the early phases of design. We integrate the capability of humans to express 3D shapes via hand-arm motions with traditional sweep surface representation to demonstrate rapid exploration of a rich variety of 3D shapes. We track the skeleton of users using the depth data provided by low-cost depth sensing camera (KinectTM). Our modeling tool is configurable to provide a variety of implicit constraints for shape symmetry and resolution based on the position, orientation and speed of the arms. An intuitive strategy for shape modifications is also proposed. Our main goal is to help the user to communicate the design intent to the computer with minimal effort. To this end, we conclusively demonstrate the creation of a wide variety of product concepts and show an average modeling time of a only few seconds while retaining the intuitiveness of the design process. ∗Address all correspondence to this author.

Handy-Potter: Rapid 3D Shape Exploration Through Natural Hand Motions

We present the paradigm of natural and exploratory shape modeling by introducing novel 3D interactions for creating, modifying and manipulating 3D shapes using arms and hands. Though current design tools provide complex modeling functionalities, they remain non-intuitive and require significant training since they segregate 3D shapes into hierarchical 2D inputs, thus binding the user to stringent procedural steps and making modifications cumbersome. In addition the designer knows what to design when they go to CAD systems and the creative exploration in design is lost. We present a shape creation paradigm as an exploration of creative imagination and externalization of shapes, particularly in the early phases of design. We integrate the capability of humans to express 3D shapes via hand-arm motions with traditional sweep surface representation to demonstrate rapid exploration of a rich variety of fairly complex 3D shapes. We track the skeleton of users using the depth data provided by low-cost depth sensing camera (Kinect™). Our modeling tool is configurable to provide a variety of implicit constraints for shape symmetry and resolution based on the position, orientation and speed of the arms. Intuitive strategies for coarse and fine shape modifications are also proposed. We conclusively demonstrate the creation of a wide variety of product concepts and show an average modeling time of a only few seconds while retaining the intuitiveness of communicating the design intent.Copyright

Estimating view parameters from random projections for Tomography using spherical MDS

BackgroundDuring the past decade, the computed tomography has been successfully applied to various fields especially in medicine. The estimation of view angles for projections is necessary in some special applications of tomography, for example, the structuring of viruses using electron microscopy and the compensation of the patients motion over long scanning period.MethodsThis work introduces a novel approach, based on the spherical multidimensional scaling (sMDS), which transforms the problem of the angle estimation to a sphere constrained embedding problem. The proposed approach views each projection as a high dimensional vector with dimensionality equal to the number of sampling points on the projection. By using SMDS, then each projection vector is embedded onto a 1D sphere which parameterizes the projection with respect to view angles in a globally consistent manner. The parameterized projections are used for the final reconstruction of the image through the inverse radon transform. The entire reconstruction process is non-iterative and computationally efficient.ResultsThe effectiveness of the sMDS is verified with various experiments, including the evaluation of the reconstruction quality from different number of projections and resistance to different noise levels. The experimental results demonstrate the efficiency of the proposed method.ConclusionOur study provides an effective technique for the solution of 2D tomography with unknown acquisition view angles. The proposed method will be extended to three dimensional reconstructions in our future work. All materials, including source code and demos, are available on https://engineering.purdue.edu/PRECISE/SMDS.

APIX: Analysis From Pixellated Inputs in Early Design Using a Pen-Based Interface

Product development is seeing a paradigm shift in the form of a simulation-driven approach. Recently, companies and designers have started to realize that simulation has the biggest impact when used as a concept verification tool in early stages of design. Early stage simulation tools like ANSYS TM Design Space and SIMULIA TM DesignSight Structure help to overcome the limitations in traditional product development processes where analyses are carried out by a separate group and not the designers. Most of these commercial tools still require well defined solid models as input and do not support freehand sketches, an integral part of the early design stage of product development. To this extent, we present APIX (acronym for Analysis from Pixellated Inputs), a tool for quick analysis of two dimensional mechanical sketches and parts from their static images using a pen-based interface. The input to the system can be offline (paper) sketches and diagrams, which include scanned legacy drawings and freehand sketches. In addition, images of twodimensional projections of three dimensional mechanical parts can also be input. We have developed an approach to extract a set of boundary contours to represent a pixellated image using known image processing algorithms. The idea is to convert the input images to online sketches and use existing stroke-based recognition techniques for further processing. The converted sketch can now be edited, segmented, recognized, merged, solved