ACM Transactions on Graphics (Proceedings SIGGRAPH Asia 2011)
Liangliang Nan1 Andrei Sharf2 Ke Xie1 Tien-Tsin Wong3 Oliver Deussen4 Daniel Cohen-Or5 Baoquan Chen1
1SIAT 2Ben Gurion Univ 3CUHK 4Konstanz Univ 5Tel Aviv Univ
Figure 1: Simplification of a complex cityscape line-drawing obtained using our Gestalt-based abstraction.
Abstract
We present a method for structural summarization and abstraction of complex spatial arrangements found in architectural drawings. The method is based on the well-known Gestalt rules, which summarize how forms, patterns, and semantics are perceived by humans from bits and pieces of geometric information. Although defining a computational model for each rule alone has been extensively studied, modeling a conjoint of Gestalt rules remains a challenge. In this work, we develop a computational framework which models Gestalt rules and more importantly, their complex interactions. We apply conjoining rules to line drawings, to detect groups of objects and repetitions that conform to Gestalt principles. We summarize and abstract such groups in ways that maintain structural semantics by displaying only a reduced number of repeated elements, or by replacing them with simpler shapes. We show an application of our method to line drawings of architectural models of various styles, and the potential of extending the technique to other computer-generated illustrations, and three-dimensional models.
Results
Figure 2: Conjoining gestalts (from Kanizsa [1980]). Overlapping (a): white dots are elements of the grid (regularity) and simultaneously belong to a curve (continuity). Conflicting (b): continuity principle of two closed curves (b-left) conflicts with the symmetry principle (b-right). Masking (c): the basis of the triangle becomes invisible as it is embedded in a group of regularly parallel lines.
Figure 6: Progressive abstraction of a complex facade based on conjoining gestalts. Zoom-ins of two different regions demonstrate our preservation of meaningful structures.
Figure 10: A sequence of abstraction steps. We color-code corresponding element groupings to visualize the computed gestalts. Two abstraction operations are performed, (a) summarization by reducing railings number in fences, and (b) embracing by replacing window elements with enclosing object. Although railings and doors overlap, their interaction is solved as railings are grouped together by regularity gestalt.
Figure 12: Comparing various abstraction techniques on the Taj Mahal (see Figure 13) . (a) shows a professional hand-drawn abstraction, (b) is a hand-drawn abstraction by an amateur artist, (c) shows state-of-the-art simplification using Shesh and Chen [2008], (d) is geometry simplification by proximity and (e) shows our gestalt-based abstraction result.
Figure 13: A sequence of gestalt based abstractions on the highly-detailed Taj Mahal.
Figure 15: Our method models the conjoining gestalts and correctly groups horizontal and vertical window structures and formations from a complex building facade.
Figure 16: Progressive gestalt abstraction of a fish mosaic. The major curve structures are identified and preserved through abstraction.
Acknowledgments
We thank the anonymous reviewers for their valuable suggestions. This work was supported in part by NSFC, 863 Program, CAS One Hundred Scholar Program, CAS Visiting Professorship for Senior International Scientists, CAS Fellowship for Young International Scientists, Shenzhen Science and Technology Foundation, China Postdoctoral Science Foundation, Israel Science Foundation, European IRG FP7, Lynn and William Frankel Center for Computer Sciences, Tuman Fund, Hong Kong RGC General Research Fund and CUHK SHIAE Project Funding.
BibTex
@ARTICLE{ConjoinGestalt2011
title = {Conjoining Gestalt Rules for Abstraction of Architectural Drawings},
author = {Liangliang Nan and Andrei Sharf and Ke Xie and Tien-Tsin Wong and Oliver Deussen and Daniel Cohen-Or and Baoquan Chen},
journal = {ACM Transactions on Graphics (Proceedings of SIGGRAPH Asia 2011)},
volume = {30},
month = {12},
pages = {185},
year = {2011},
}
