Inverse Procedural Modeling of Branching Structures by Inferring L-Systems

Jianwei Guo; Haiyong Jiang; Bedrich Benes; Oliver Deussen; Xiaopeng Zhang; Dani Lischinski; Hui Huang

doi:10.1145/3394105

Inverse Procedural Modeling of Branching Structures by Inferring L-Systems

Jianwei Guo, Haiyong Jiang, Bedrich Benes, Oliver Deussen, Xiaopeng Zhang, Dani Lischinski, Hui Huang

The Rachel and Selim Benin School of Engineering and Computer Science

Research output: Contribution to journal › Article › peer-review

29 Scopus citations

Abstract

We introduce an inverse procedural modeling approach that learns L-system representations of pixel images with branching structures. Our fully automatic model generates a compact set of textual rewriting rules that describe the input. We use deep learning to discover atomic structures such as line segments or branchings. Orientation and scaling of these structures are determined and the detected structures are combined into a tree. The initial representation is analyzed, and repeating parts are encoded into a small grammar by using greedy optimization while the user can control the size of the detected rules. The output is an L-system that represents the input image as a simple text and a set of terminal symbols. We apply our approach to a variety of examples, demonstrate its robustness against noise and blur, and we show that it can detect user sketches and complex input structures.

Original language	American English
Article number	3394105
Journal	ACM Transactions on Graphics
Volume	39
Issue number	5
DOIs	https://doi.org/10.1145/3394105
State	Published - Sep 2020

Bibliographical note

Funding Information:
This work was supported in parts by National Key R&D Program (2018YFB2100602), NSFC (61861130365, 61761146002, 61802406, 61802362), GD Higher Education Key Program (2018KZDXM058), LHTD (20170003), GD Leading Talent Program (00201509), DFG (422037984), NSF (10001387), FAR (602757), and GD Laboratory of Artificial Intelligence and Digital Economy (SZ). Authors’ addresses: J. Guo and X. Zhang, NLPR, Institute of Automation, CAS, Beijing, China; emails: jianwei.guo@nlpr.ia.ac.cn, Xiaopeng.Zhang@ia.ac.cn; H. Jiang, UCAS, Beijing, NTU Singapore; email: haiyong.jiang1990@gmail.com; B. Benes, Pur-due University; email: bbenes@purdue.edu; O. Deussen, SIAT Shenzhen and University Konstanz; email: oliver.deussen@uni-konstanz.de; D. Lischinski, The Hebrew University of Jerusalem; email: danix3d@gmail.com; H. Huang (corresponding author), College of Computer Science & Soft-ware Engineering, Shenzhen University; email: hhzhiyan@gmail.com. Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. Copyrights for components of this work owned by others than ACM must be honored. Abstracting with credit is permitted. To copy otherwise, or republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee. Request permissions from permissions@acm.org. © 2020 Association for Computing Machinery. 0730-0301/2020/06-ART155 $15.00 https://doi.org/10.1145/3394105

Publisher Copyright:
© 2020 ACM.

Keywords

L-systems
grammar induction
procedural generation

Access to Document

10.1145/3394105

Cite this

@article{d422acf8638347c9a33417d4524248d0,

title = "Inverse Procedural Modeling of Branching Structures by Inferring L-Systems",

abstract = "We introduce an inverse procedural modeling approach that learns L-system representations of pixel images with branching structures. Our fully automatic model generates a compact set of textual rewriting rules that describe the input. We use deep learning to discover atomic structures such as line segments or branchings. Orientation and scaling of these structures are determined and the detected structures are combined into a tree. The initial representation is analyzed, and repeating parts are encoded into a small grammar by using greedy optimization while the user can control the size of the detected rules. The output is an L-system that represents the input image as a simple text and a set of terminal symbols. We apply our approach to a variety of examples, demonstrate its robustness against noise and blur, and we show that it can detect user sketches and complex input structures.",

keywords = "L-systems, grammar induction, procedural generation",

author = "Jianwei Guo and Haiyong Jiang and Bedrich Benes and Oliver Deussen and Xiaopeng Zhang and Dani Lischinski and Hui Huang",

note = "Funding Information: This work was supported in parts by National Key R&D Program (2018YFB2100602), NSFC (61861130365, 61761146002, 61802406, 61802362), GD Higher Education Key Program (2018KZDXM058), LHTD (20170003), GD Leading Talent Program (00201509), DFG (422037984), NSF (10001387), FAR (602757), and GD Laboratory of Artificial Intelligence and Digital Economy (SZ). Authors{\textquoteright} addresses: J. Guo and X. Zhang, NLPR, Institute of Automation, CAS, Beijing, China; emails: jianwei.guo@nlpr.ia.ac.cn, Xiaopeng.Zhang@ia.ac.cn; H. Jiang, UCAS, Beijing, NTU Singapore; email: haiyong.jiang1990@gmail.com; B. Benes, Pur-due University; email: bbenes@purdue.edu; O. Deussen, SIAT Shenzhen and University Konstanz; email: oliver.deussen@uni-konstanz.de; D. Lischinski, The Hebrew University of Jerusalem; email: danix3d@gmail.com; H. Huang (corresponding author), College of Computer Science & Soft-ware Engineering, Shenzhen University; email: hhzhiyan@gmail.com. Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. Copyrights for components of this work owned by others than ACM must be honored. Abstracting with credit is permitted. To copy otherwise, or republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee. Request permissions from permissions@acm.org. {\textcopyright} 2020 Association for Computing Machinery. 0730-0301/2020/06-ART155 $15.00 https://doi.org/10.1145/3394105 Publisher Copyright: {\textcopyright} 2020 ACM.",

year = "2020",

month = sep,

doi = "10.1145/3394105",

language = "American English",

volume = "39",

journal = "ACM Transactions on Graphics",

issn = "0730-0301",

publisher = "Association for Computing Machinery (ACM)",

number = "5",

}

TY - JOUR

T1 - Inverse Procedural Modeling of Branching Structures by Inferring L-Systems

AU - Guo, Jianwei

AU - Jiang, Haiyong

AU - Benes, Bedrich

AU - Deussen, Oliver

AU - Zhang, Xiaopeng

AU - Lischinski, Dani

AU - Huang, Hui

N1 - Funding Information: This work was supported in parts by National Key R&D Program (2018YFB2100602), NSFC (61861130365, 61761146002, 61802406, 61802362), GD Higher Education Key Program (2018KZDXM058), LHTD (20170003), GD Leading Talent Program (00201509), DFG (422037984), NSF (10001387), FAR (602757), and GD Laboratory of Artificial Intelligence and Digital Economy (SZ). Authors’ addresses: J. Guo and X. Zhang, NLPR, Institute of Automation, CAS, Beijing, China; emails: jianwei.guo@nlpr.ia.ac.cn, Xiaopeng.Zhang@ia.ac.cn; H. Jiang, UCAS, Beijing, NTU Singapore; email: haiyong.jiang1990@gmail.com; B. Benes, Pur-due University; email: bbenes@purdue.edu; O. Deussen, SIAT Shenzhen and University Konstanz; email: oliver.deussen@uni-konstanz.de; D. Lischinski, The Hebrew University of Jerusalem; email: danix3d@gmail.com; H. Huang (corresponding author), College of Computer Science & Soft-ware Engineering, Shenzhen University; email: hhzhiyan@gmail.com. Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. Copyrights for components of this work owned by others than ACM must be honored. Abstracting with credit is permitted. To copy otherwise, or republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee. Request permissions from permissions@acm.org. © 2020 Association for Computing Machinery. 0730-0301/2020/06-ART155 $15.00 https://doi.org/10.1145/3394105 Publisher Copyright: © 2020 ACM.

PY - 2020/9

Y1 - 2020/9

N2 - We introduce an inverse procedural modeling approach that learns L-system representations of pixel images with branching structures. Our fully automatic model generates a compact set of textual rewriting rules that describe the input. We use deep learning to discover atomic structures such as line segments or branchings. Orientation and scaling of these structures are determined and the detected structures are combined into a tree. The initial representation is analyzed, and repeating parts are encoded into a small grammar by using greedy optimization while the user can control the size of the detected rules. The output is an L-system that represents the input image as a simple text and a set of terminal symbols. We apply our approach to a variety of examples, demonstrate its robustness against noise and blur, and we show that it can detect user sketches and complex input structures.

AB - We introduce an inverse procedural modeling approach that learns L-system representations of pixel images with branching structures. Our fully automatic model generates a compact set of textual rewriting rules that describe the input. We use deep learning to discover atomic structures such as line segments or branchings. Orientation and scaling of these structures are determined and the detected structures are combined into a tree. The initial representation is analyzed, and repeating parts are encoded into a small grammar by using greedy optimization while the user can control the size of the detected rules. The output is an L-system that represents the input image as a simple text and a set of terminal symbols. We apply our approach to a variety of examples, demonstrate its robustness against noise and blur, and we show that it can detect user sketches and complex input structures.

KW - L-systems

KW - grammar induction

KW - procedural generation

UR - http://www.scopus.com/inward/record.url?scp=85091985900&partnerID=8YFLogxK

U2 - 10.1145/3394105

DO - 10.1145/3394105

M3 - Article

AN - SCOPUS:85091985900

SN - 0730-0301

VL - 39

JO - ACM Transactions on Graphics

JF - ACM Transactions on Graphics

IS - 5

M1 - 3394105

ER -

Inverse Procedural Modeling of Branching Structures by Inferring L-Systems

Abstract

Bibliographical note

Keywords

Access to Document

Other files and links

Fingerprint

Cite this