SBM09: Sketch Based Interfaces and Modeling 2009

Permanent URI for this collection

https://diglib.eg.org/handle/10.2312/417

Browse

Now showing 1 - 20 of 20

Automatic Evaluation of Sketch Recognizers
(The Eurographics Association, 2009) Schmieder, Paul; Plimmer, Beryl; Blagojevic, Rachel; Cindy Grimm and Joseph J. LaViola, Jr.
We present our toolkit to automatically evaluate recognition algorithms. There are few published comparative evaluations of sketch recognition algorithms and those that exist do not provide benchmarking or direct comparisons because standardised data and an evaluation platform is not available. By unifying data collection, labelling and evaluation in one tool, fair, flexible and comprehensive evaluations are possible. Currently we have 6 existing recognizers integrated into this tool. With our initial evaluations of these recognizers we have observed that the context from which training data is taken has an effect on recognition success rates. These results suggest that an evaluation platform such as this is a powerful adjunct for sketch recognition research.
A descriptor for large scale image retrieval based on sketched feature lines
(The Eurographics Association, 2009) Eitz, Mathias; Hildebrand, Kristian; Boubekeur, Tamy; Alexa, Marc; Cindy Grimm and Joseph J. LaViola, Jr.
We address the problem of large scale sketch based image retrieval, searching in a database of over a million images. The search is based on a descriptor that elegantly addresses the asymmetry between the binary user sketch on the one hand and the full color image on the other hand. The proposed descriptor is constructed such that both the full color image and the sketch undergo exactly the same preprocessing steps. We also design an adapted version of the descriptor proposed for MPEG-7 and compare their performance on a database of 1.5 million images. Best matching images are clustered based on color histograms, to offset the lacking color in the query. Overall, the query results demonstrate that the system allows users an intuitive access to large image databases.
Editing Level-Set Models with Sketched Curves
(The Eurographics Association, 2009) Eyiyurekli, M.; Grimm, C.; Breen, D.; Cindy Grimm and Joseph J. LaViola, Jr.
Level set models are deformable implicit surfaces where the deformation is controlled by a speed function in the level set partial differential equation (PDE). These models are widely used in computer graphics applications due to their implicit definition, low-level volumetric representation and the powerful numerical techniques used to produce the PDE-based deformation. We present a set of interactive sketch-based level-set surface editing operators. These operators allow a user to sketch curves above or on a level-set surface in order to edit the surface's shape. Once the curves are sketched the surface interactively evolves to locally fit to the curves. A user may then modify the curves in order to refine the shape of the model. The mathematics, algorithms and techniques needed to implement numerous sketch-based level set modeling capabilities are described. The speed functions that produce the surface deformations within the context of solving the level-set PDE are detailed. Several examples are presented to demonstrate the flexibility and usefulness of the editing operators.
The Effect of Task on Classification Accuracy: Using Gesture Recognition Techniques in Free-Sketch Recognition
(The Eurographics Association, 2009) Field, Martin; Gordon, Sam; Peterson, Eric; Robinson, Raquel; Stahovich, Thomas; Alvarado, Christine; Cindy Grimm and Joseph J. LaViola, Jr.
Generating, grouping, and labeling free-sketch data is a difficult and time-consuming task for both user study participants and researchers. To simplify this process for both parties, we would like to have users draw isolated shapes instead of complete sketches that must be hand-labeled and grouped, and then use this data to train our free-sketch symbol recognizer. However, it is an open question whether shapes draw in isolation accurately reflect the way users draw shapes in a complete diagram. Furthermore, many of the simplest shape recognition algorithms were designed to recognize gestures, and it is not clear that they will generalize to freely-drawn shapes. To answer these questions, we perform experiments using three different recognizers to measure the effect of the data collection task on recognition accuracy. We find that recognizers trained only on isolated shapes can classify freely-sketched shapes as well as the same recognizers trained on free-sketches. We also show that user-specific training examples significantly improve recognition rates. Finally, we introduce a variant of a popular and simple gesture recognition algorithm that recognizes freely-drawn shapes as well as a highly-accurate but more complex recognizer designed explicitly for free-sketch recognition.
Games For Sketch Data Collection
(The Eurographics Association, 2009) Johnson, Gabe; Do, Ellen Yi-Luen; Cindy Grimm and Joseph J. LaViola, Jr.
This article describes sketching games made for the purpose of collecting data about how people make and describe hand-made drawings. The approach leverages human computation, whereby players provide information about drawings in exchange for entertainment. The games facilitate the collection of raw sketch input and associates it with human-provided text descriptions. Researchers may browse and download this data for their own purposes such as training sketch recognizers. Two systems with distinct game mechanics are described: Picturephone and Stellasketch. The system architectures are briefly presented, followed by a discussion of our initial results using sketching games as a research platform for sketch recognition and interaction.
Modeling from Contour Drawings
(The Eurographics Association, 2009) Kraevoy, Vladislav; Sheffer, Alla; Panne, Michiel van de; Cindy Grimm and Joseph J. LaViola, Jr.
Occlusion contours are a natural feature to draw when tracing an object in an image or when drawing an object. We investigate the development of 3D models from multi-stroke contour drawings with the help of a 3D template model that serves as a shape prior. The template is aligned and then deformed by our method to match the drawn contours. At the heart of this process is the need to provide good correspondences between points on the contours and vertices on the model, which we pose as an optimisation problem using a hidden Markov model. An alternating correspond-and-deform process then progressively deforms the 3D template to match the image contours. We demonstrate the method on a wide range of examples.
Multi-touch Focus+Context Sketch-based Interaction
(The Eurographics Association, 2009) Hahne, Uwe; Schild, Jonas; Elstner, Stefan; Alexa, Marc; Cindy Grimm and Joseph J. LaViola, Jr.
In this paper we present a Focus+Context screen for combined pen and touch interaction. A tabletop display presents contextual information and enables multi-touch detection for navigation through frustrated total internal reflection (FTIR). A high resolution pen enabled display is continuously localized on the tabletop. It is used as a movable focus display and pen input device. The resulting system has a large virtual resolution for both display and pen interaction. We demonstrate the advantages of Focus+Context for pen-based interaction in combination with multi-touch navigation in applications that make use of sketch-based interfaces.
On Expert Performance in 3D Curve-Drawing Tasks
(The Eurographics Association, 2009) Schmidt, Ryan; Khan, Azam; Kurtenbach, Gord; Singh, Karan; Cindy Grimm and Joseph J. LaViola, Jr.
A study is described which examines the drawing accuracy of experts when drawing foreshortened projections of 3D curves in ecologically-valid conditions. The main result of this study is that the distribution of error in expert drawings exhibits a bias similar to that previously observed in non-expert subjects, which is dependent on the degree of foreshortening of the imagined drawing surface. A review of existing perceptual studies also finds that only absolute 2D image-space error has been considered, which has been found to be largest with viewing angles of 25-55. Our visualizations of 3D error indicate that 3D bias continues to increase with decreasing viewing angle. Based on these findings, we analyze current 3D curve drawing techniques for susceptibility to foreshortening bias, and make some suggestions for future sketch-based modeling systems.
Revisiting ShortStraw Improving Corner Finding in Sketch-Based Interfaces
(The Eurographics Association, 2009) Xiong, Yiyan; Jr., Joseph J. LaViola; Cindy Grimm and Joseph J. LaViola, Jr.
We present IStraw, a new corner finding technique based on an analysis of the ShortStraw algorithm. Our analysis reveals several limitations in ShortStraw and we develop techniques to overcome them. We also present an extension to our corner finding approach for dealing with ink strokes that contain curves and arcs. An evaluation of our approach shows significant accuracy improvements over ShortStraw for polyline ink strokes with and without curves using an all-or-nothing accuracy metric while still maintaining ShortStraw's computational complexity.
Shadow Buttons: Exposing WIMP Functionality While Preserving the Inking Surface in Sketch-Based Interfaces
(The Eurographics Association, 2009) Marinkas, Diane; Zeleznik, Robert C.; Jr., Joseph J. LaViola; Cindy Grimm and Joseph J. LaViola, Jr.
We present Shadow Buttons, an approach to placing WIMP interface elements into gestural and sketch-based interfaces. Utilizing the hover state, supported by pen-based devices such as Tablet PCs, we provide users with important WIMP-based functionality by invoking widgets only when the stylus hovers over "shadow" regions. In this way, every pixel of the display, including the shadows, can be drawn on. By interacting with a shadow region while in the hover state, users can temporarily display and interact with familiar WIMP elements. We explore the Shadow Button design space as it relates to handwritten mathematical expressions and also present an informal evaluation of our technique, examining various Shadow Button parameters including button size, placement, and method of invocation. Preliminary results indicate that users prefer utilizing the hover state over tapping on the Shadow Button and in general, prefer interface elements to be as close to the ink as possible. In addition, the distance between the shadow region and the menu was found to be the most important factor in Shadow Button usability.
Sketch-Based Interaction and Calligraphic Tags to Create Comics Online
(The Eurographics Association, 2009) Lopes, Ricardo; Cardoso, Tiago; Silva, Nelson; Fonseca, Manuel J.; Cindy Grimm and Joseph J. LaViola, Jr.
Over the last years, amateur artists have been using the new Internet technologies to easily produce, share and distribute their comics. However, web applications for this kind of creation are still restricting users by adopting simple and very limiting interaction methods. Therefore, these systems do not allow the creation of visually rich comics and do not support some of the inherent needs of comics, such as the repetition of previous elements. In this paper, we propose a solution for creating comics online which builds upon the traditional principles of creating paper comics. Our approach combines a sketch-based interaction to draw comics with rich edition possibilities, computer assisted drawing techniques and a calligraphic retrieval mechanism for reusing previous elements. Experimental evaluation showed that this approach is better suited for these problems than existent applications, and that users can create more appealing and richer comics with higher flexibility and efficiency.
A Sketch-based Interface for Photo Pop-up
(The Eurographics Association, 2009) Ventura, J.; DiVerdi, S.; Höllerer, T.; Cindy Grimm and Joseph J. LaViola, Jr.
We present sketch-based tools for single-view modeling which allow for quick 3D mark-up of a photograph. With our interface, detailed 3D models can be produced quickly and easily. After establishing the background geometry, foreground objects can be cut out using our novel sketch-based segmentation tools. These tools make use of the stroke speed and length to help determine the user's intentions. Depth detail is added to the scene by drawing occlusion edges. Such edges play an important part in human scene understanding, and thus provide an intuitive form of input to the modeling system. Initial results and evaluation show that our methods produce good 3D results in a short amount of time and with little user effort, demonstrating the usefulness of an intelligent sketching interface for this application domain.
Sketch-Based Subdivision Models
(The Eurographics Association, 2009) Nasriy, A.; Karam, W. Bou; Samavati, F.; Cindy Grimm and Joseph J. LaViola, Jr.
Designing a control mesh (or a polyhedron) for a subdivision model is a tedious task. It involves many difficult decisions such as how to minimize the number of extraordinary vertices, how best to choose their valencies, and where to place them in the control mesh. In this paper, we present an intuitive and interactive approach for using sketch-based interface to design subdivision models. The input to the system is a set of strokes forming the profile curves of the surface. From the constructed control polygons of the sketched curves, a coarse and quad dominant control mesh is generated with few extraordinary vertices or faces. The corresponding limit surface interpolates the profile curves with the capability of local control across these curves and of the model in general. Although our approach is oriented towards quad-based systems such as Catmull-Clark, it could well be adopted in other subdivision schemes.
Sketching Subdivision Surfaces
(The Eurographics Association, 2009) Bein, M.; Havemann, S.; Stork, A.; Fellner, Dieter W.; Cindy Grimm and Joseph J. LaViola, Jr.
We describe a 3D modeling system that combines subdivision surfaces with sketch-based modeling in order to meet two conflicting goals: ease of use and fine-grained shape control. For the excellent control, low-poly modeling is still the method of choice for creating high-quality 3D models, e.g., in the games industry. However, direct mesh editing can be very tedious and time consuming. Our idea is to include also stroke-based techniques for rapidly modeling regular surface parts. We propose a simple and efficient algorithm for converting a 2D stroke to a control polygon suitable for Catmull/Clark subdivision surfaces. We have realized a small but reasonably rich set of interactive modeling tools to assess the expressiveness of stroke-based mesh design with a number of examples.
Sort, Merge, Repeat: An Algorithm for Effectively Finding Corners in Hand-sketched Strokes
(The Eurographics Association, 2009) Wolin, A.; Paulson, B.; Hammond, T.; Cindy Grimm and Joseph J. LaViola, Jr.
Free-sketch recognition systems attempt to recognize freely-drawn sketches without placing stylistic constraints on the users. Such systems often recognize shapes by using geometric primitives that describe the shape's appearance rather than how it was drawn. A free-sketch recognition system necessarily allows users to draw several primitives using a single stroke. Corner finding, or vertex detection, is used to segment these strokes into their underlying primitives (lines and arcs), which in turn can be passed to the geometric recognizers. In this paper, we present a new multi-pass corner finding algorithm called MergeCF that is based on continually merging smaller stroke segments with similar, larger stroke segments in order to eliminate false positive corners. We compare MergeCF to two benchmark corner finders with substantial improvements in corner detection.
Tools for the Efficient Generation of Hand-Drawn Corpora Based on Context-Free Grammars
(The Eurographics Association, 2009) MacLean, Scott; Tausky, David; Labahn, George; Lank, Edward; Marzouk, Mirette; Cindy Grimm and Joseph J. LaViola, Jr.
In sketch recognition systems, ground-truth data sets serve to both train and test recognition algorithms. Unfortunately, generating data sets that are sufficiently large and varied is frequently a costly and time-consuming endeavour. In this paper, we present a novel technique for creating a large and varied ground-truthed corpus for hand drawn math recognition. Candidate math expressions for the corpus are generated via random walks through a context-free grammar, the expressions are transcribed by human writers, and an algorithm automatically generates ground-truth data for individual symbols and inter-symbol relationships within the math expressions. While the techniques we develop in this paper are illustrated through the creation of a ground-truthed corpus of mathematical expressions, they are applicable to any sketching domain that can be described by a formal grammar.
Towards Beautification of Freehand Sketches using Suggestions
(The Eurographics Association, 2009) Murugappan, S.; Sellamani, S.; Ramani, K.; Cindy Grimm and Joseph J. LaViola, Jr.
Beautification of freehand sketches is integral for building robust sketch understanding systems and sketch-based inter-faces 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 am-biguous 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.
A Usability Evaluation of AlgoSketch: A Pen-Based Application for Mathematics
(The Eurographics Association, 2009) O'Connell, Theresa; Li, Chuanjun; Miller, Timothy S.; Zeleznik, Robert C.; Jr., Joseph J. LaViola; Cindy Grimm and Joseph J. LaViola, Jr.
AlgoSketch is a pen-based system for entering and editing mathematics. It supports interactive computation and early-stage mathematical algorithm design. In its paper-like environment, mathematical expressions can be en- tered anywhere on the page. Recognition and computational feedback are given in real time. We present results of a formative user evaluation of AlgoSketch, examining its applicability as a new interaction paradigm and users' overall experience with its mathematical entry, feedback, and computational support. Using a goal, question met- ric (GQM) framework, we evaluated AlgoSketch using efficiency, effectiveness, and satisfaction metrics. Logging data was supported by a post-questionnaire and anecdotal data. Results indicate acceptance of the AlgoSketch paradigm; strong potential for workplace utility, and a need for better mathematical expression recognition.
Visual Languages and Visual Thinking: Sketch Based Interaction and Modeling
(The Eurographics Association, 2009) Gross, M. D.; Cindy Grimm and Joseph J. LaViola, Jr.
Research on sketching with computers dates to the earliest days of modern computing. Recent work in this area, combined with other advances in hardware and software technologies promises, finally, significant impact. The kinds, qualities, and purposes of sketch-based interaction, or visual languages, vary as widely as do other forms of language. In addition to practical applications in every domain, advances in sketch-based interaction and modeling can help us understand and support visual thinking.
VolumeViewer: An Interactive Tool for Fitting Surfaces to Volume Data
(The Eurographics Association, 2009) Sowell, R.; Liu, L.; Ju, T.; Grimm, C.; Abraham, C.; Gokhroo, G.; Low, D.; Cindy Grimm and Joseph J. LaViola, Jr.
Recent advances in surface reconstruction algorithms [BM07,LBD08] allow surfaces to be built from contours lying on non-parallel planes. Such algorithms allow users to construct surfaces of similar quality more efficiently by using a small set of oblique contours, rather than many parallel contours. However, current medical imaging systems do not provide tools for sketching contours on oblique planes. In this paper, we take the first steps towards bridging the gap between the new surface reconstruction technologies and putting those methods to use in practice. We develop a novel interface for modeling surfaces from volume data by allowing the user to sketch contours on arbitrarily oriented cross-sections of the volume, and we examine the users' ability to contour the same structures using oblique cross-sections with similar consistency as they can using parallel cross-sections. We measure the inter-observer and intra-observer variability of trained physicians contouring on oblique cross-sections of real patient data as compared to the traditional parallel cross-sections, and show that the variation is much higher for oblique contouring. We then show that this variability can be greatly reduced by integrating a collection of training images into the interface.

Browse

Browsing SBM09: Sketch Based Interfaces and Modeling 2009 by Title

Results Per Page

Sort Options