Stroke Correspondence Construction Using Manifold Learning

Abstract

Stroke correspondence construction is a precondition for generating inbetween frames from a set of key frames. In our case, each stroke in a key frame is a vector represented as a Disk B-Spline Curve (DBSC) which is a flexible and compact vector format. However, it is not easy to construct correspondences between multiple DBSC strokes effectively because of the following points: (1) with the use of shape descriptors, the dimensionality of the feature space is high; (2) the number of strokes in different key frames is usually large and different from each other and (3) the length of corresponding strokes can be very different. The first point makes matching difficult. The other two points imply 'many to many' and 'part to whole' correspondences between strokes. To solve these problems, this paper presents a DBSC stroke correspondence construction approach, which introduces a manifold learning technique to the matching process. Moreover, in order to handle the mapping between unequal numbers of strokes with different lengths, a stroke reconstruction algorithm is developed to convert the 'many to many' and 'part to whole' stroke correspondences to 'one to one' compound stroke correspondence.

Direct dynamics study of the hydrogen abstraction reaction of CF3CH2Cl + Cl → CF3CHCl + HCl

“…Here, the single‐point energy calculations are based on the B3LYP/6‐311+G(2d,2p) and MP2/6‐311G(d,p) optimized geometries, respectively. It has been shown that in many studies 17–21 this method can provide reliable PES information. Original G3(MP2) calculations are also performed for comparison.…”

Vector graphics animation with time-varying topology

“…Early stroke-based approaches [Burtnyk and Wein 1971;Reeves 1981;Fekete et al 1995] are manual (the animator selects pairs of strokes to interpolate) and topology-unaware (strokes are interpolated independently, unaware of their neighbors). Recent methods [Liu et al 2011;Yu et al 2012] use shape descriptors and machine learning techniques to compute stroke correspondences automatically, but since they are topologyunaware, they are unable to generate space-time continuous animation with time-varying topology.…”

FTP‐SC: Fuzzy Topology Preserving Stroke Correspondence

“…In [LCY*11, YBS*12], a stroke reconstruction algorithm is proposed to resolve the topology ambiguities. However, such algorithm often fails in practice, since it is akin to a difficult artificial intelligence problem to infer the occluded parts from the key drawings or to associate the stroke(s) that correspond to a common feature between the successive key frames.…”