Optimal Mass Transport for Shape Matching and Comparison

Zhengyu Su, Yalin Wang, Rui Shi, Wei Zeng, Jian Sun, Feng Luo, Xianfeng Gu

Research output: Contribution to journalArticle

22 Citations (Scopus)

Abstract

Surface based 3D shape analysis plays a fundamental role in computer vision and medical imaging. This work proposes to use optimal mass transport map for shape matching and comparison, focusing on two important applications including surface registration and shape space. The computation of the optimal mass transport map is based on Monge-Brenier theory, in comparison to the conventional method based on Monge-Kantorovich theory, this method significantly improves the efficiency by reducing computational complexity from O(n2) to O(n). For surface registration problem, one commonly used approach is to use conformal map to convert the shapes into some canonical space. Although conformal mappings have small angle distortions, they may introduce large area distortions which are likely to cause numerical instability thus resulting failures of shape analysis. This work proposes to compose the conformal map with the optimal mass transport map to get the unique area-preserving map, which is intrinsic to the Riemannian metric, unique, and diffeomorphic. For shape space study, this work introduces a novel Riemannian framework, Conformal Wasserstein Shape Space, by combing conformal geometry and optimal mass transport theory. In our work, all metric surfaces with the disk topology are mapped to the unit planar disk by a conformal mapping, which pushes the area element on the surface to a probability measure on the disk. The optimal mass transport provides a map from the shape space of all topological disks with metrics to the Wasserstein space of the disk and the pullback Wasserstein metric equips the shape space with a Riemannian metric. We validate our work by numerous experiments and comparisons with prior approaches and the experimental results demonstrate the efficiency and efficacy of our proposed approach.

Original languageEnglish (US)
Article number7053911
Pages (from-to)2246-2259
Number of pages14
JournalIEEE Transactions on Pattern Analysis and Machine Intelligence
Volume37
Issue number11
DOIs
StatePublished - Nov 1 2015

Fingerprint

Optimal Transport
Shape Space
Shape Matching
Mass Transport
Mass transfer
Conformal Map
Shape Analysis
Conformal Mapping
Riemannian Metric
Registration
Conformal mapping
Wasserstein Metric
Conformal Geometry
Metric
Transport Theory
Numerical Instability
3D shape
Pullback
Medical Imaging
Computer Vision

Keywords

  • optimal mass transport
  • shape representation
  • shape space
  • surface matching

ASJC Scopus subject areas

  • Artificial Intelligence
  • Computer Vision and Pattern Recognition
  • Software
  • Computational Theory and Mathematics
  • Applied Mathematics

Cite this

Optimal Mass Transport for Shape Matching and Comparison. / Su, Zhengyu; Wang, Yalin; Shi, Rui; Zeng, Wei; Sun, Jian; Luo, Feng; Gu, Xianfeng.

In: IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol. 37, No. 11, 7053911, 01.11.2015, p. 2246-2259.

Research output: Contribution to journalArticle

Su, Zhengyu ; Wang, Yalin ; Shi, Rui ; Zeng, Wei ; Sun, Jian ; Luo, Feng ; Gu, Xianfeng. / Optimal Mass Transport for Shape Matching and Comparison. In: IEEE Transactions on Pattern Analysis and Machine Intelligence. 2015 ; Vol. 37, No. 11. pp. 2246-2259.
@article{30627140feee4144a81c3938de9c6550,
title = "Optimal Mass Transport for Shape Matching and Comparison",
abstract = "Surface based 3D shape analysis plays a fundamental role in computer vision and medical imaging. This work proposes to use optimal mass transport map for shape matching and comparison, focusing on two important applications including surface registration and shape space. The computation of the optimal mass transport map is based on Monge-Brenier theory, in comparison to the conventional method based on Monge-Kantorovich theory, this method significantly improves the efficiency by reducing computational complexity from O(n2) to O(n). For surface registration problem, one commonly used approach is to use conformal map to convert the shapes into some canonical space. Although conformal mappings have small angle distortions, they may introduce large area distortions which are likely to cause numerical instability thus resulting failures of shape analysis. This work proposes to compose the conformal map with the optimal mass transport map to get the unique area-preserving map, which is intrinsic to the Riemannian metric, unique, and diffeomorphic. For shape space study, this work introduces a novel Riemannian framework, Conformal Wasserstein Shape Space, by combing conformal geometry and optimal mass transport theory. In our work, all metric surfaces with the disk topology are mapped to the unit planar disk by a conformal mapping, which pushes the area element on the surface to a probability measure on the disk. The optimal mass transport provides a map from the shape space of all topological disks with metrics to the Wasserstein space of the disk and the pullback Wasserstein metric equips the shape space with a Riemannian metric. We validate our work by numerous experiments and comparisons with prior approaches and the experimental results demonstrate the efficiency and efficacy of our proposed approach.",
keywords = "optimal mass transport, shape representation, shape space, surface matching",
author = "Zhengyu Su and Yalin Wang and Rui Shi and Wei Zeng and Jian Sun and Feng Luo and Xianfeng Gu",
year = "2015",
month = "11",
day = "1",
doi = "10.1109/TPAMI.2015.2408346",
language = "English (US)",
volume = "37",
pages = "2246--2259",
journal = "IEEE Transactions on Pattern Analysis and Machine Intelligence",
issn = "0162-8828",
publisher = "IEEE Computer Society",
number = "11",

}

TY - JOUR

T1 - Optimal Mass Transport for Shape Matching and Comparison

AU - Su, Zhengyu

AU - Wang, Yalin

AU - Shi, Rui

AU - Zeng, Wei

AU - Sun, Jian

AU - Luo, Feng

AU - Gu, Xianfeng

PY - 2015/11/1

Y1 - 2015/11/1

N2 - Surface based 3D shape analysis plays a fundamental role in computer vision and medical imaging. This work proposes to use optimal mass transport map for shape matching and comparison, focusing on two important applications including surface registration and shape space. The computation of the optimal mass transport map is based on Monge-Brenier theory, in comparison to the conventional method based on Monge-Kantorovich theory, this method significantly improves the efficiency by reducing computational complexity from O(n2) to O(n). For surface registration problem, one commonly used approach is to use conformal map to convert the shapes into some canonical space. Although conformal mappings have small angle distortions, they may introduce large area distortions which are likely to cause numerical instability thus resulting failures of shape analysis. This work proposes to compose the conformal map with the optimal mass transport map to get the unique area-preserving map, which is intrinsic to the Riemannian metric, unique, and diffeomorphic. For shape space study, this work introduces a novel Riemannian framework, Conformal Wasserstein Shape Space, by combing conformal geometry and optimal mass transport theory. In our work, all metric surfaces with the disk topology are mapped to the unit planar disk by a conformal mapping, which pushes the area element on the surface to a probability measure on the disk. The optimal mass transport provides a map from the shape space of all topological disks with metrics to the Wasserstein space of the disk and the pullback Wasserstein metric equips the shape space with a Riemannian metric. We validate our work by numerous experiments and comparisons with prior approaches and the experimental results demonstrate the efficiency and efficacy of our proposed approach.

AB - Surface based 3D shape analysis plays a fundamental role in computer vision and medical imaging. This work proposes to use optimal mass transport map for shape matching and comparison, focusing on two important applications including surface registration and shape space. The computation of the optimal mass transport map is based on Monge-Brenier theory, in comparison to the conventional method based on Monge-Kantorovich theory, this method significantly improves the efficiency by reducing computational complexity from O(n2) to O(n). For surface registration problem, one commonly used approach is to use conformal map to convert the shapes into some canonical space. Although conformal mappings have small angle distortions, they may introduce large area distortions which are likely to cause numerical instability thus resulting failures of shape analysis. This work proposes to compose the conformal map with the optimal mass transport map to get the unique area-preserving map, which is intrinsic to the Riemannian metric, unique, and diffeomorphic. For shape space study, this work introduces a novel Riemannian framework, Conformal Wasserstein Shape Space, by combing conformal geometry and optimal mass transport theory. In our work, all metric surfaces with the disk topology are mapped to the unit planar disk by a conformal mapping, which pushes the area element on the surface to a probability measure on the disk. The optimal mass transport provides a map from the shape space of all topological disks with metrics to the Wasserstein space of the disk and the pullback Wasserstein metric equips the shape space with a Riemannian metric. We validate our work by numerous experiments and comparisons with prior approaches and the experimental results demonstrate the efficiency and efficacy of our proposed approach.

KW - optimal mass transport

KW - shape representation

KW - shape space

KW - surface matching

UR - http://www.scopus.com/inward/record.url?scp=84960913566&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84960913566&partnerID=8YFLogxK

U2 - 10.1109/TPAMI.2015.2408346

DO - 10.1109/TPAMI.2015.2408346

M3 - Article

VL - 37

SP - 2246

EP - 2259

JO - IEEE Transactions on Pattern Analysis and Machine Intelligence

JF - IEEE Transactions on Pattern Analysis and Machine Intelligence

SN - 0162-8828

IS - 11

M1 - 7053911

ER -