arrow
返回

StripMaker: Perception-driven Learned Vector Sketch Consolidation

delete2023-07-26
delete3
PRE
AI
C
Chenxi Liu *
T
Toshiki Aoki
M
Mikhail Bessmeltsev
A
Alla Sheffer
DOI:10.1145/3592130delete
delete原文链接
delete原文求助
delete分享
delete收藏
摘要

摘要

En 中文
Artist sketches often use multiple overdrawn strokes to depict a single intended curve. Humans effortlessly mentally consolidate such sketches by detecting groups of overdrawn strokes and replacing them with the corresponding intended curves. While this mental process is near instantaneous, manually annotating or retracing sketches to communicate this intended mental image is highly time consuming; yet most sketch applications are not designed to handle overdrawing and can only operate on overdrawing-free, consolidated sketches. We propose StripMaker, a new and robust learning based method for automatic consolidation of raw vector sketches. We avoid the need for an unsustainably large manually annotated learning corpus by leveraging observations about artist workflow and perceptual cues viewers employ when mentally consolidating sketches. We train two perceptionaware classifiers that assess the likelihood that a pair of stroke groups jointly depicts the same intended curve: our first classifier is purely local and only accounts for the properties of the evaluated strokes; our second classifier incorporates global context and is designed to operate on approximately consolidated sketches. We embed these classifiers within a consolidation framework that leverages artist workflow: we first process strokes in the order they were drawn and use our local classifier to arrive at an approximate consolidation output, then use the contextual classifier to refine this output and finalize the consolidated result. We validate StripMaker by comparing its results to manual consolidation outputs and algorithmic alternatives. StripMaker achieves comparable performance to manual consolidation. In a comparative study participants preferred our results by a 53% margin over those of the closest algorithmic alternative (67% versus 14%, other/neither 19%).
Keyword:
Vector graphics
sketch consolidation
line art

期刊

ACM Transactions on Graphics 封面图
ACM Transactions on Graphics
IF:
9.5
论文数:
4.7K
被引数:
3.6W

机构

U
University of Tokyo
学者数:
7.1W
论文数: 6.5W
被引数: 2.2K
U
universite de montreal
学者数:
4.6W
论文数: 3.8W
被引数: 46
U
University of British Columbia
学者数:
7.0W
论文数: 6.1W
被引数: 8.6W
学者 查看更多机构
引用论文

引用论文

OpenSketch: A Richly-Annotated Dataset of Product Design Sketches
err2019-11-08
err57
errOAAI
errGryaditskaya, Yulia; Sypesteyn, Mark; Hoftijzer, Jan Willem; Pont, Sylvia; Durand, Fredo; Bousseau, Adrien
err分享
err收藏
A Benchmark for Rough Sketch Cleanup
err2020-11-27
err32
errOAAI
errYan, Chuan; Vanderhaeghe, David; Gingold, Yotam
err分享
err收藏
Spontaneous and Mitogen-induced Activity of Lymphocytes of Different Density in Multiple Sclerosis
err2008-01-30
err0
PREAI
errC.J.J. Brinkman; W.M. Nillesen; O.R. Hommes; K.J.B. Lamers; B.E. de Pauw
err分享
err收藏
err分享
err收藏
Endpoint fusing method of online freehand-sketched polyhedrons
err2018-11-07
err4
PREAI
errWang, Shuxia; Zhang, Qian; Wang, Shouxia; Jing, Xiaoke; Gao, Mantun
err分享
err收藏
Handling gaps for vector graphics coloring
err2021-07-15
err4
PREAI
errJiang, Jie; Seah, Hock Soon; Liew, Hong Ze
err分享
err收藏
学者 查看更多内容