A critical component of diagramming sketch tools is their ability to reliably recognise hand-drawn diagram components. This is made difficult by the presence of both geometric shapes and characters in diagrams. The goal of our research is to improve sketch recognition by improving the accuracy in grouping and classifying strokes in a diagram into text characters and shapes. We have done this by identifying the most significant features of strokes that can be used to distinguish shapes from text using a decision tree based partitioning technique. Implementation and evaluation of this new “shape divider” using these features against InkKit’s existing divider and the Microsoft divider has shown that our divider is more accurate at dividing text and shape strokes and can therefore improve overall sketch recognition.

Exploring better techniques for diagram recognition

Sketch recognition is one of the essential step of sketch understanding. Challenge in sketch recognition is the variation and imprecision present in sketch. Free drawing styles of sketching make it difficult to build a robust sketch recognition system. This paper proposes a novel recognition approach that can recognize primitive shapes, as well as combinations of these primitives. The approach is independent of stroke order, number, as well as invariant to size and aspect ratio of sketch. Feature string is used to represent primitives. We defined a similarity measure on these feature strings that counts common substrings in two input strings, which is referred to as the string kernel in the field of kernel methods. Support vector machine(SVM) is then trained with labeled examples to handle the task of classification. The experiment on hand drawn digital circuit diagrams shows that our system can recognize sketching efficiently and robustly.

Sketch recognition via string kernel

Within the product development process, working in virtual environments is generally applied. Thus, the digital product model in its different stages is an essential prerequisite for the virtual modeling and simulation of manufacturing processes. Product development starts with conceptual design including the concepts for shape and function. The conceptual designers work on fuzzy product data applied to perception modeling and shape creation. For some time now, there has been a strong demand to define creative human problem solving in conceptual design as a task of virtual product development and to assist the process simultaneously through the use of appropriate digital models and virtual methods. This paper reports on R&D results concerned with integration of the process of conceptual design based on traditional aids, such as freehand paper sketches, in virtual environments. Until now, it has not been possible to work iteratively with both freehand sketches and digital product models. This paper describes how to interpret product descriptions in freehand sketches and to transfer them to digital three-dimensional models, e.g. in order to visualize, calculate and evaluate shapes to the development of computer-aided design (CAD) models. The methods combine an image-based modeling strategy with conventional CAD modeling strategies.

The implementation of a sketch-based virtual product development

The virtual product data model in its different stages is an essential prerequisite for the virtual modeling of processes and their virtual simulation Product design starts with conceptual design including the concepts for the future productpsilas shape and function - the gestalt In this stage, the designers work on fuzzy product data interpreted in so-called scribbles and applied to perception modeling and shape creating In the paper we will suppose methods to analyze paper scribbles for the creation of fuzzy-spline pre-CAD shape models In particular, the paper discusses special geometrical issues that are concerned with the interpretation of the future productpsilas contours within paper scribbles and virtual pre-CAD models Examples of curve and closed region modeling strategies based on a blurred 2D description will be presented The implemented algorithms combine image-based modeling methods with usual computer-aided design (CAD) modeling strategies

Chapter 11: Creative Shape Interpretation in Paper Scribbles

Presentation, in a form of a sketched graphical diagram, is often used as a training approach of various didactic levels. However, the trainee examination using the hand-sketch diagrams is still not a common practice, due to the evaluation complexity; both for a teacher and for an automatic recognition system. The elaborated interface, supported by the low-level recognition algorithm, simplifies the evaluation procedures. The solution removes both interpretation uncertainty factor and simple errors from the sketch, using the user interaction interface. The users interaction is based on online pattern recognition of strokes in a colour context. Their representation is returned to the user and stored as a graph structure. Additionally, the interface tracks the sketch creation process for more precise final evaluation. The interface can support many fields as various as business, computer science, mathematics, engineering, music and so on; where a set of graphical symbols are lines and curves composition.

The colour sketch recognition interface for training systems

Sketch recognition is an essential process for sketch understanding. The free drawing style of sketching makes it difficult to build a robust sketch recognition system that can support the imprecision and high variability present in sketch. This paper addresses these problems inherent in sketch recognition in the framework of Bayesian network which can readily represent uncertainty in the recognition process and make inference based on partial evidence. To further improve recognition accuracy, context information is incorporated in the recognition process rather than identifying sketches in isolation. The new framework proposed in this paper offers user more sketching freedom by automatically grouping strokes and recognizing editing gestures as well as over-traced strokes.

An incremental Bayesian approch to sketch recognition [approach read approach]

Sketch recognition is an essential process for Sketch Understanding. The free drawing style of sketching makes it difficult to build a robust sketch recognition system that can support the imprecision and high variability present in sketch. This paper addresses these problems inherent in sketch recognition in the framework of Bayesian network which can readily represent uncertainty in the recognition process and make inference based on partial evidence. To further improve recognition accuracy, context information is incorporated in the recognition process rather than identifying sketches in isolation. The new framework proposed in this paper offers user more sketching freedom by automatically grouping strokes and recognizing editing gestures as well as over-traced strokes.

http://ieeexplore.ieee.org/iel5/10231/32629/01527740.pdf

Xiao-Jun Wang

Papers

An incremental Bayesian approch to sketch recognition [approach read approach]

An incremental bayesian approch to sketch recognition