FTP‐SC: Fuzzy Topology Preserving Stroke Correspondence

Abstract

Stroke correspondence construction is a precondition for vectorized 2D animation inbetweening and remains a challenging problem. This paper introduces the FTP‐SC, a fuzzy topology preserving stroke correspondence technique, which is accurate and provides the user more effective control on the correspondence result than previous matching approaches. The method employs a two‐stage scheme to progressively establish the stroke correspondence construction between the keyframes. In the first stage, the stroke correspondences with high confidence are constructed by enforcing the preservation of the so‐called “fuzzy topology” which encodes intrinsic connectivity among the neighboring strokes. Starting with the high‐confidence correspondences, the second stage performs a greedy matching algorithm to generate a full correspondence between the strokes. Experimental results show that the FTP‐SC outperforms the existing approaches and can establish the stroke correspondence with a reasonable amount of user interaction even for keyframes with large geometric and spatial variations between strokes.

A benchmark for rough sketch cleanup

“…2D-to-3D lifting algorithms often assume that continuity and junctions in the 2D artwork imply continuity and junctions in the 3D shape [Andre and Saito 2011;Bessmeltsev et al 2015;Kaplan and Cohen 2006;Lipson and Shpitalni 1996;Xu et al 2014;Zheng et al 2016]. In-betweening algorithms similarly assume that continuity and junctions should be preserved during interpolation [Whited et al 2010;Yang et al 2018 Fig. 2.…”

“…In the future, we would like to explore end-to-end evaluations on specific downstream sketch processing tasks like 3D reconstruction [Xu et al 2014] or animation in-betweening [Whited et al 2010;Yang et al 2018]. We would also like to resolve imperfections in human-created ground truth related to junctions and stroke thickness and color.…”

“…In contrast, the later stages are time-consuming, tedious, and costly. Decades of research into sketch processing have produced a large literature of algorithms for tasks such as 3D inference [Andre and Saito 2011;Bessmeltsev et al 2015;Kaplan and Cohen 2006;Lipson and Shpitalni 1996;Shao et al 2012Shao et al , 2013Shtof et al 2013;Xu et al 2014;Zheng et al 2016], UI design [Landay and Myers 1995], and in-between animation [Whited et al 2010;Yang et al 2018]. These research works remain underused in practice since they typically expect clean, sketch-like input rather than the rough, messy sketches found in the wild.…”

Stroke‐Based Drawing and Inbetweening with Boundary Strokes

“…[CMB17], Yang et al . [YSC*18] for relevant technology. As strokes are corresponded, the occlusion in each individual frame can be handled independently based on the frame's boundary strokes as described in Section 4.…”

Stroke Correspondence by Labeling Closed Areas

“…This system enables users to manually specify several constraints and handle the estimated results (correspondence). In addition, Yang et al [19] extend Yang's method [11] to establish characteristics of stroke connections, and improve the estimation accuracy. However, with these methods, the users must manually redraw keyframes with vector lines by tracing over the original character im-age in advance.…”