SBM10: Sketch Based Interfaces and Modeling 2010

Permanent URI for this collection

https://diglib.eg.org/handle/10.2312/416

Browse

Now showing 1 - 20 of 20

Sketching Variational Hermite-RBF Implicits
(The Eurographics Association, 2010) Brazil, Emilio Vital; Macedo, Ives; Sousa, Mario Costa; Figueiredo, Luiz Henrique de; Velho, Luiz; Marc Alexa and Ellen Yi-Luen Do
We present techniques for modeling Variational Hermite Radial Basis Function (VHRBF) Implicits using a set of sketch-based interface and modeling (SBIM) operators. VHRBF Implicits is a simple and compact representation well suited for SBIM. It provides quality reconstructions, preserving the intended shape from a coarse and nonuniform number of point-normal samples extracted directly from the input strokes. In addition, it has a number of desirable properties such as parameter-free modeling, invariance under geometric similarities on the input strokes, suitable estimation of differential quantities, good behavior near close sheets, and both linear fitting and reproduction. Our approach uses these properties of VHRBF Implicits to quickly and robustly generate the overall shape of 3D models. We present examples of implicit models obtained from a set of SBIM language operators for contouring, cross-editing, kneading, oversketching and merging.
Exploring Usability and Learnability of Mode Inferencing in Pen/Tablet Interfaces
(The Eurographics Association, 2010) Negulescu, Matei; Ruiz, Jaime; Lank, Edward; Marc Alexa and Ellen Yi-Luen Do
The inferred mode protocol uses contextual reasoning and local mediators to eliminate the need to access specific modes to perform draw, select, move and delete operations in a sketch interface. In this paper, we describe an observational experiment to understand the learnability, user preference and frequency of use of mode inferencing in a sketch application. The experiment demonstrated that those participants instructed in the interface features liked the fluid transitions between modes. As well, interaction techniques were not self-revealing: Participants who were not instructed in interaction techniques took longer to learn about inferred mode features and were more negative about the interaction techniques. Over multiple sketching sessions, as users develop expertise with the system, we find that they combine inferred mode techniques to speed interaction, and frequently make use of scratch space on the display to retrain themselves and to tune their behaviors. Our results inform the design of sketch interface techniques that incorporate noncommand features.
A Pen-Based Tool for Visualizing Vector Mathematics
(The Eurographics Association, 2010) Bott, Jared N.; Jr., Joseph J. LaViola; Marc Alexa and Ellen Yi-Luen Do
We present VectorPad, a novel, pen-based application for three-dimensional vector mathematics visualization. VectorPad allows users to define vectors and perform mathematical operations on them through the recognition of handwritten mathematics. The user interface consists of a sketching area, where the user can write vector definitions and operations such as addition, subtraction, scalar multiplication, and cross product, and a 3D graph for visualization. Vectors are visualized dynamically on the graph, which can be manipulated by the user. We also performed a short, informal user study evaluating the user interface and visualizations of VectorPad. Results from the study show that visualizations were generally well liked but the application needs to provide a more comprehensive set of visualization tools as well as refinement to some of the animations.
Folding Avoidance in Skeletal Strokes
(The Eurographics Association, 2010) Asente, Paul J.; Marc Alexa and Ellen Yi-Luen Do
Skeletal strokes is a method for mapping vector artwork defined along a straight path onto an arbitrary destination path. It is commonly used to give sketches a hand-drawn look, by mapping a scanned natural-medium stroke onto a user-drawn path. When the destination path turns corners or has areas with a small radius of curvature, the mapped artwork takes on an undesirable folded or creased appearance inside the corner or curve. We present a method to eliminate the folds in many cases by adjusting the skeleton before using it to map the artwork. While this cannot address all cases with folding, it gives greatly improved results in the cases that it can address.
Computing Confidence Values for Geometric Constraints for use in Sketch Recognition
(The Eurographics Association, 2010) Johnston, Joshua; Hammond, Tracy; Marc Alexa and Ellen Yi-Luen Do
Geometric constraints are used by many sketch recognition systems to perform high-level assembly of components of a sketch into semantic structures. However, with a few notable exceptions, most of the current recognition systems do not have constraints that use real-valued notions of confidence. We discuss methods for assigning confidence values to different kinds of constraints. We show how these confidence values equate to user perception, how they can be used to balance speed and accuracy in recognition algorithms, and how they can be used to assign confidence values to the high-level shapes they are used to construct. We use these constraints to extend the LADDER shape definition language in a system that recognizes 5,900 hand-drawn examples of 485 different military course-of-action diagrams at an accuracy of 89.9%.
Sketch-Based Modeling of Vascular Systems: a First Step Towards Interactive Teaching of Anatomy
(The Eurographics Association, 2010) Pihuit, Adeline; Cani, Marie-Paule; Palombi, Olivier; Marc Alexa and Ellen Yi-Luen Do
We present a sketch-based modeling system, inspired from anatomical drawing, which constructs plausible 3D models of branching vessels from a single sketch. The input drawing typically includes non-flat silhouettes and occluded parts. We exploit the sketching conventions used in anatomical drawings to infer depth and curvature from contour and skeleton curves extracted from the sketch. We then model the set of branching vessels as a convolution surface generated by a graph of skeleton curves: while these curves are set to fit the sketch in the front plane, non-uniform B-spline interpolation is used to give them smoothly varying depth values that meet the set of constraints. The final model is displayed using an expressive rendering method that imitates the aspect of chalk drawing. We discuss the future use of this system as a step towards the interactive teaching of anatomy.
Drawing with the Flow: A Sketch-Based Interface for Illustrative Visualization of 2D Vector Fields
(The Eurographics Association, 2010) Schroeder, David; Coffey, Dane; Keefe, Dan; Marc Alexa and Ellen Yi-Luen Do
This paper presents Drawing with the Flow, a sketch-based interface for illustrating 2D vector fields. Drawing with the Flow explores the problem of making scientific visualization tools accessible to artists and illustrators, who have a finely tuned visual design sense, but do not typically have the programming or mathematical background required to work with modern visualization algorithms and tools. Using a sketch-based interface that is accessible to illustrations, Drawing with the Flow makes it possible for illustrators to explore new visual designs for streamline placement in order to create illustrations of 2D vector fields, such as simulated fluid flows. The interface includes a method for interpreting hand-drawn marks relative to an underlying vector field, utilizing an ink-data settling procedure to subtly maintain an image consistent with the underlying data. A variable-density interactive streamline seeding algorithm tied to sketch-based input is also introduced, and design lessons learned and limitations are discussed. Several example illustration results have been created during sessions ranging from five to twenty-five minutes. The results presented demonstrate different styles of illustration for describing simulated data for 2D flows past a cylinder and 2D flows that include several different types of critical points.
Sketch-Based Recognition System for General Articulated Skeletal Figures
(The Eurographics Association, 2010) Zamora, Shane; Sherwood, Timothy; Marc Alexa and Ellen Yi-Luen Do
We present a new recognition system for detecting general articulated skeletal figures in sketch-based applications. We abstract drawing style from recognition by defining figures using two data models: templates and figure targets. Our system recognizes general skeletal figures consisting of lines and ellipses which may be drawn partially off the canvas with the system automatically completing the figure. Figures are modeled as graphs and are allowed to contain cycles. Subgraph matching on a graph built from the stroke input is used to perform recognition. This paper outlines our system design, key details to its proper implementation, and proposes its application to various domains of sketch recognition.
Exploring Frame Gestures for Fluid Freehand Sketching
(The Eurographics Association, 2010) Nijboer, Menno; Gerl, Moritz; Isenberg, Tobias; Marc Alexa and Ellen Yi-Luen Do
In this paper, we explore a minimalistic, gesture-based interface for fluid freehand concept sketching with vector graphics. Our approach leverages the advantages of both the GUI and gestural interface paradigms. We describe how to use frame gestures to control rotation, translation, and scale of the drawing canvas and of stroke selections. Based on an implementation of this concept we evaluate our tool with both novices and experts, and report on both its benefits and drawbacks.
Shape Modeling with Sketched Feature Lines in Immersive 3D Environments
(The Eurographics Association, 2010) Perkunder, Helen; Israel, Johann Habakuk; Alexa, Marc; Marc Alexa and Ellen Yi-Luen Do
In this paper we address the question whether automatic shape creation using sketched strokes as input in an immersive 3D environment supports the sketching process in early phases of product design. To investigate this question, model creation and deformation algorithms of the desktop sketch-based modeling tool FiberMesh [NISA07] were transferred to an immersive 3D environment. A comparative user study was conducted among twelve design students and professional designers. Line-based sketching in a 3D environment and sketch-based modeling, both in a 3D and 2D environment were compared. The analysis of the study yielded few differences between the conditions, but two findings were made: usability for a creative sketching task was perceived higher for line-based sketching than for sketch-based modeling - both in an immersive 3D environment. Shape modeling in immersive 3D environments was perceived as more stimulating and attractive than under 2D conditions.
Assisted Multitouch Image-Based Reconstruction
(The Eurographics Association, 2010) Bauer, Frank; Stamminger, Marc; Marc Alexa and Ellen Yi-Luen Do
We present an image-based reconstruction approach for mobile, multitouch enabled devices. A novel scene description based on a multi-agent system is used to allow a real-time reconstruction workflow even on devices with relatively low processing speeds. Using the built in camera along with data from an accelerometer, compass and GPS module allows us to easily add new camera objects to our reconstruction world with an initial estimation for position and orientation. Our multi-agent scene description proved to be flexible enough to perform modelling tasks beyond image-based reconstruction using multitouch gestures only.
Modeling 2.5D Plants from Ink Paintings
(The Eurographics Association, 2010) Amati, Cristina; Brostow, Gabriel J.; Marc Alexa and Ellen Yi-Luen Do
Chinese ink painting, known as sumi-e, is a traditional art form based on careful placement of expressive brush strokes. To extend such brush strokes from their paper form, we propose a system that can digitize and transform such paintings into 2:5D sprites, while preserving the artist's unique style. We apply our paintstroke-extraction technique to plants, the most traditional subject of sumi-e, and one where CG modeling is fairly hard, but procedural or simulated animation techniques can subsequently be brought to bear. Instead of forcing artists to use pressure-sensitive digital tablets, we address the problem of how to non-invasively decompose a real painting of a plant into semantically meaningful components, like leaves and petals. Webcam filming of the artist at work provides valuable cues. This part of our system needs no human interaction and no prior training, and achieves its goal solely by analyzing the video timeline to find even intersecting brush strokes. Afterwards, it constructs a model of the plant and textures it with the extracted brush strokes, inpainting when necessary.
The Power of Automatic Feature Selection: Rubine on Steroids
(The Eurographics Association, 2010) Blagojevic, Rachel; Chang, Samuel Hsiao-Heng; Plimmer, Beryl; Marc Alexa and Ellen Yi-Luen Do
Digital ink features drive recognition engines. Intuitively, we understand that particular features are of more value for some problems than others. Likewise, inclusion of poor features may be detrimental to recognition success. Many different ink features have been proposed for ink recognition, and most work well for the context that they are employed. However given a new problem it is not clear which of the already defined features will be most useful. We have assembled and categorized a comprehensive feature library and use this with attribute selection algorithms to choose the best features for a specified problem. To verify the effectiveness of this approach the selected features are used to train a Rubine's recognizer. We show that a set of complementary features is most effective: poor features adversely affect recognition as do two or more aliases of good features. We have composed a variant of a Rubine recognizer for 3 different datasets and compared these with the Rubine's original features, a variant on this InkRubine and $1. The results show that feature selection can significantly improve recognition rates with this simple algorithm thus verifying our hypothesis that the right combination of features for a problem is one key to recognition success.
Stroke Extraction and Classification for Mesh Inflation
(The Eurographics Association, 2010) Olsen, Luke; Samavati, Faramarz F.; Marc Alexa and Ellen Yi-Luen Do
We provide a method for extracting and classifying stroke segments from a line drawing or sketch with the goal of producing perceptually-valid output in the context of mesh inflation. This is important as processing freehand sketch input is a fundamental task in sketch-based interfaces, yet many systems bypass the problem by forcing simplified, unnatural drawing patterns. Our stroke extraction combines contour tracing with feature-preserving post-processing. The extracted strokes are classified according to the objects and regions in the sketch: object and region boundaries, interior features, and suggestive lines. The outcome of this classification is demonstrated with examples in feature-sensitive mesh inflation.
A Sketching Interface for Sitting-Pose Design
(The Eurographics Association, 2010) Lin, Juncong; Igarashi, Takeo; Mitani, Jun; Saul, Greg; Marc Alexa and Ellen Yi-Luen Do
We present a sketch interface for interactively placing a 3D human character in a sitting position on a chair. The user first sketches the target pose as a 2D stick figure. The user can specify whether a joint will be attached to the environment (for example, the feet may be put on the ground) with a pin tool. Our system then reconstructs the 3D pose from the sketch figure considering the constraints specified by the user and the interaction between the character and the environment. This paper presents a user interface and a reconstruction algorithm that combines a genetic algorithm and a quasi-Newton solver to efficiently find a collision-free pose. An informal user study showed the effectiveness of our approach.
Rata.SSR: Data Mining for Pertinent Stroke Recognizers
(The Eurographics Association, 2010) Chang, Samuel Hsiao-Heng; Plimmer, Beryl; Blagojevic, Rachel; Marc Alexa and Ellen Yi-Luen Do
While many approaches to digital ink recognition have been proposed, most lack flexibility and adaptability to provide acceptable recognition rates across a variety of problem spaces. Time and expert knowledge are required to build accurate recognizers for a new domain. This project uses selected algorithms from a data mining toolkit and a large feature library, to compose a tailored software component (Rata.SSR) that enables single stroke recognizer generation from a few example diagrams. We evaluated Rata.SSR against four popular recognizers with three data sets (one of our own and two from other projects). The results show that it outperforms other recognizers on all tests except recognizer and data set pairs (e.g. PaleoSketch recognizer and PaleoSketch data set) in these cases the difference is small, and Rata is more flexible. We hence demonstrate a flexible and adaptable procedure for adopting existing techniques to quickly generate accurate recognizers without extensive knowledge of either AI or data mining.
Investigating the Learnability of Immersive Free-Hand Sketching
(The Eurographics Association, 2010) Wiese, Eva; Israel, Johann Habakuk; Meyer, Achim; Bongartz, Sara; Marc Alexa and Ellen Yi-Luen Do
Immersive modeling systems which allow for sketching and constructing three-dimensional product models receive growing interest from both academic research and industry [DBWB*00; FASM02; IWMS09; KAMD*08; KAML*01]. The potential of such systems is estimated in the possibility to create models in a one-to-one scale, to interact with product models and to reduce breaks between analogue and digital media during the development process [DBWB*00; IWMS09]. Despite the growing interest in 3D-sketching techniques, little is known so far about the ability of designers to create free-hand drawings in three-dimensional space. In particular, only few studies have investigated cognitive and sensorimotor processes during immersive sketching yet. This paper contributes to the research on immersive sketching by investigating the learnability of free-hand sketching in an experimental setting. In a study among 25 students, participants repeatedly sketched primitive shapes (circles, squares, balls, and cubes). Sketching performance was operationalized by the time needed to complete a sketch, the quality of the sketch, and the subjective mental workload of the designers. Results suggest a significant enhancement of sketching quality over time, but no change in the time needed to complete a sketch.
Modeling-in-Context: User Design of Complementary Objects with a Single Photo
(The Eurographics Association, 2010) Lau, Manfred; Saul, Greg; Mitani, Jun; Igarashi, Takeo; Marc Alexa and Ellen Yi-Luen Do
The products that we use everyday are typically designed and produced for mass consumption. However, it is difficult for such products to satisfy the needs of individual users. We present a framework that allows the end-user to participate in the entire process of designing their own objects, from the initial concept stage to the production of a new real-world object that fits well with the existing complementary objects. We advocate using a single photo as a rough guide for the user to sketch a new customized object that does not exist in the photo. Our system provides a 2D interface for sketching the outline of the new object and annotating certain geometric properties of it directly on the photo. We introduce a Modified Lipson optimization method for generating the 3D shape. We design a variety of real-world everyday objects that are complementary to the existing objects and environment in the photo. We show that novice users can learn and create new objects with our system within minutes.
Inkus: A Freehand Method of Creating Business Process Models
(The Eurographics Association, 2010) Mangano, Nicolas; Sukaviriya, Noi; Marc Alexa and Ellen Yi-Luen Do
Business process modeling has become an industry wide practice for business transformation consulting. Tools that support business process modeling are designed for experienced users to draw a process with precision and professional appearance. These tools are not conducive to sketching quick business design ideas. In this paper, we explore a non-intrusive business process sketching tool which allows free hand sketches of process ideas and slowly brings the users to the required common business vocabulary. We also explore adding business values that support process design thinking, along with sketching. Our goal is to help unleash creativity in business designers and enrich the design process with values beyond drawing. This paper presents a design concept and an implemented prototype of such a system.
Feature Extraction and Classifier Combination for Image-based Sketch Recognition
(The Eurographics Association, 2010) Tumen, R. Sinan; Acer, M. Emre; Sezgin, T. Metin; Marc Alexa and Ellen Yi-Luen Do
Image-based approaches to sketch recognition typically cast sketch recognition as a machine learning problem. In systems that adopt image-based recognition, the collected ink is generally fed through a standard three stage pipeline consisting of the feature extraction, learning and classification steps. Although these approaches make regular use of machine learning, existing work falls short of presenting a proper treatment of important issues such as feature extraction, feature selection, feature combination, and classifier fusion. In this paper, we show that all these issues are significant factors, which substantially affect the ultimate performance of a sketch recognition engine. We support our case by experimental results obtained from two databases using representative sets of feature extraction, feature selection, classification, and classifier combination methods. We present the pros and cons of various choices that can be made while building sketch recognizers and discuss their trade-offs.

Browse

Browsing SBM10: Sketch Based Interfaces and Modeling 2010 by Issue Date

Results Per Page

Sort Options