Hierarchical parsing and recognition of hand-sketched diagrams

Kara, Levent Burak; Stahovich, Thomas F.

doi:10.1145/1281500.1281528

Cited by 33 publications

(39 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Then, random noise and affine distortions were applied to these 15 flow chart symbol classes to generate a total of 3000 symbols. The 3000 flow chart symbols were divided into training and testing sets of 1000 and 2000 symbols, respectively, For the testing data set, the average classification accuracy of VizDraw is found to be 98.7%, while the classification accuracy of SimuSketch [15] is found to be 94% when these symbols were used. Further the performance of VizDraw is compared with SVM-HMM hybrid approach [17] and traditional HMM approach on a limited data set consisting of five flowchart symbols.…”

Section: Resultsmentioning

confidence: 99%

See 1 more Smart Citation

VizDraw: A Platform to Convert Online Hand-Drawn Graphics into Computer Graphics

Mishra

Eichel

Fieguth

et al. 2009

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Abstract. With the adoption of tablet-based data entry devices, there is considerable interest in methods for converting hand-drawn sketches of flow charts, graphs and block diagram into accurate machine interpretations, a conversion process with many applications in engineering, presentations, and simulations. However, the recognition of hand-drawn graphics is a great challenge due to the visual similarity of many system components. This is complicated due to the significant differences in drawing styles between users. The proposed method, VizDraw, establishes an architecture that utilizes a number of pattern recognition tools to convert hand-drawn diagrams into computer graphics by segmenting the original diagram into individual components. This method generates hypothesis graphs for each component, evaluates the hypotheses using forward and backward dynamic programming, and finally utilizes a rule-based floor planning routine for component and symbol placement. VizDraw is invariant to scaling, rotation, translation and style of drawing. The preliminary results show how VizDraw is used for engineering drawings, simulation, and incorporation into computer aided design (CAD) models.

show abstract

Section: Resultsmentioning

confidence: 99%

“…These include hidden Markov models [14][15][16][17][18], Bayesian networks [9], neural networks [19], and wavelet networks [20,21].…”

Section: Introductionmentioning

confidence: 99%

VizDraw: A Platform to Convert Online Hand-Drawn Graphics into Computer Graphics

Mishra

Eichel

Fieguth

et al. 2009

Lecture Notes in Computer Science

View full text Add to dashboard Cite

show abstract

“…This is different to "on demand" recognition where the ink is only recognised when the user activates a widget. Kara and Stahovich (2004) have used "on demand" recognition in their system as the results of the recognition may interrupt the user during the design process. However, all the strokes must be sorted beforehand before intelligent ink editing can be accomplished.…”

Section: Ink Recognitionmentioning

confidence: 99%

Intelligent mind-mapping

Chik

Plimmer

Hosking

2007

Proceedings of the 19th Australasian Conference on Computer-Human Interaction: Entertaining User Interfaces

View full text Add to dashboard Cite

“…4. There are many approaches to recognition, which can be arranged in several main categories. There are image-based approaches, like the one given by Kara [15]. The general problem here is that segmentation and clustering cannot be done at all.…”

Section: Introductionmentioning

confidence: 99%

A model-based recognition engine for sketched diagrams

Brieler

Minas

2010

Journal of Visual Languages & Computing

View full text Add to dashboard Cite

Many of today's recognition approaches for hand-drawn sketches are feature-based, which is conceptually similar to the recognition of hand-written text. While very suitable for the latter (and more tasks, e.g., for entering commands), such approaches do not easily allow for clustering and segmentation of strokes, which is crucial to their recognition. This results in applications which do not feel natural but impose artificial restrictions on the user regarding how diagrams and single components are to be drawn.In this paper we propose a recognition approach based on models, which is designed for the mentioned issue of clustering and segmentation. All strokes are fed into different models, where each model is responsible for a certain type of primitive, e.g., a line or an arc. The recognition of a component in the drawing is then decomposed into the recognition of its primitives, which can be directly searched for in the models. Finally, the identified primitives are assembled to the complete component.In several case studies we also show the applicability and generality of our approach, as very different types of components can be recognized. Furthermore, the proposed approach is part of a complete system to sketch understanding which can not only recognize single components, but can also reason about diagrams as a whole, consisting of a set of these components.

show abstract

Hierarchical parsing and recognition of hand-sketched diagrams

Cited by 33 publications

References 14 publications

VizDraw: A Platform to Convert Online Hand-Drawn Graphics into Computer Graphics

VizDraw: A Platform to Convert Online Hand-Drawn Graphics into Computer Graphics

Intelligent mind-mapping

A model-based recognition engine for sketched diagrams

Contact Info

Product

Resources

About