Automated freehand sketch segmentation using radial basis functions

Abstract

Freehand sketching is widely regarded as an efficient and natural way for interaction between computers and humans. We present a robust computerized scheme to automatically segment freehand sketches into a series of components with specific geometric meaning regardless of whether these are generated online or offline. This task is a necessary first step toward sketch understanding. By exploiting the interpolation / extrapolation characteristic of Radial Basis Functions (RBFs), a greedy algorithm consisting of forward and backward operations is proposed for finding the minimum set of segmentation points that can be used to reconstruct with high fitting accuracy freehand sketches in the form of implicit functions. To obtain segmentation points, a simple angle based rule is used to remove "bridging" points that provide a smooth transition between consecutive sketch components. Feasibility of the proposed algorithm is demonstrated by a preliminary performance assessment study using ten computer generated drawings. These experiments show that in this dataset sensitivity of the segmentation was higher than 97.5% with a false positive (FP) rate of approximately 25%. The majority of false positive identifications are located on arc regions where a larger number of segmentation points are needed for reconstruction purposes. The primary contribution of this algorithm is that it transforms an ambiguous problem namely, freehand sketch segmentation, into an implicit function fitting operation. Therefore, this proposed approach has several advantages including independence of the actual sketching activity, and the ability for a satisfactory detection of the transition point between a line and an arc or between two arcs.

A new paradigm based on agents applied to free-hand sketch recognition

“…On the one hand, the "segmentation" of the stroke is carried out extracting the primitive geometrical forms which make it up and providing information about them which will be used in the decision stages afterwards. This segmentation process, which is currently a well-known problem in the sketch recognition field (Pu & Gur, 2009;Company, Varley, Piquer, Vergara, & Sanchez-Rubio, 2009) On the other hand, a series of features or basic "cues" of the introduced stroke are extracted. Only features that are not dependent on scale, position and orientation have been chosen, like the FFT (Tao, Morrow, Heinemann, & Sommer, 1995) of the radius signature (also called polar signature) and of the "arc length versus cumulative turning angle" signature (also known as direction signature).…”

Human Perception in Segmentation of Sketches

“…As a result, a new approach may be considered good simply because it finds the ideal number of segmentation points, even though their actual locations are far from ideal (see, for example, [4]). …”

“…Segmentation of sketches is an open problem in the process of sketch recognition. One recent contribution can be found in [4]. We can note, in passing, that even segmentation of line drawings remains an unsolved problem.…”

“…In future, we intend to fit the sketches to parametric curves (for example, by applying some variant of the approach described in [4]), and use natural coordinates for coordinates and statistical values.…”

Recognition of sketching from surface electromyography

“…It is well-known that sketching is a key mode of informal, perceptual interaction that has been shown to be especially valuable for creative design tasks [1,2,3,4]. For designers, the ability to rapidly sketch objects with uncertain types, sizes, shapes, and positions is important to the creative process [1].…”