10th IAPR-TC15 Workshop on

Graph-based Representations in

Pattern Recognition

May 13-15, 2015

Beijing, China

Sponsored by: 

1

Welcome Message

Welcome to the 10th IAPR-TC15 Workshop on Graph-based Representations in Pattern

Recognition (GbR2015) in Beijing, China, 2015! The series of GbR Workshop is sponsored

by the TC15 (Graph-based Representations) of IAPR (International Association for Pattern

Recognition), started from 1997. The GbR2015 follows the editions of 1997 in Lyon (France),

1999 in Castle of Haindorf (Austria), 2001 in Ischia (Italy), 2003 in York (UK), 2005 in

Poitiers (France), 2007 in Alicante (Spain), 2009 in Venice (Italy), 2011 in Münster

(Germany), and 2013 in Vienna (Austria).

Graph is a very effective structure for representing structural patterns and has been widely

used in pattern recognition, computer vision, machine learning, and data mining. Structural

pattern recognition methods usually represent a pattern as a graph modeling the primitives

and their interrelationship. A digital image can be viewed as a graph with its pixels as nodes.

Many machine learning methods view sample points as the nodes of a graph for modeling

their affinity. Social media and biological data are often represented as graphs or networks.

Theory and methods of graph data analysis have been studied intensively for graph generation,

matching, labeling, segmentation, clustering, classification, and data mining in various fields.

The GbR Workshop encourages all related works from such fields.

The GbR2015 received 53 full submissions. The authors are from Asia, Europe, America,

and Oceania. Each submission was assigned to three Program Committee (PC) members for

review, and 11 sub-reviewers were invited by the PC members. Based on the reviews, 36

papers were accepted for presentation at the workshop and inclusion in the proceedings. The

accepted papers cover diverse issues of graph-based methods and applications, with 7 in

graph representation, 15 in graph matching, 7 in graph clustering and classification, and 7 in

graph-based applications.

In addition to regular presentations, we invited two established researchers to present

keynote speeches. Professor Marcello Pelillo of University of Venice (Italy) and Dr. Xing Xie

of Microsoft Research Asia (Beijing, China) will present speeches titled “Revealing Structure

in Large Graphs: Szemerdi’s Regularity Lemma and Its Use in Pattern Recognition” and

“Understanding Users by Connecting Large Scale Social Graphs,” respectively. We are

grateful to the invited speakers, the authors of all submitted papers, the Program Committee

members, and reviewers, and the staff of the Organizing Committee. Without their

contributions, this workshop would not have been a success. We also thank Springer for

publishing the proceedings, especially Mr. Alfred Hofmann and Ms. Anna Kramer for their

efforts and patience in collecting and editing the proceedings.

We welcome you to take part in this academic event and hope you find GbR2015 an

enjoyable and fruitful workshop.

May 2015

Cheng-Lin Liu

Bin Luo

Walter G. Kropatsch

Jian Cheng

2

Organizing Committee

Sponsors

National Laboratory of Pattern Recognition, Institute of Automation of CAS

IAPR TC-15

Program Co-Chairs

Cheng-Lin Liu Institute of Automation of CAS, China

Bin Luo Anhui University, China

Walter G. Kropatsch Vienna University of Technology, Austria

Technical Program Committee:

Xiang Bai Huazhong University of Science and Technology, China

Xiao Bai Beihang University, China

Roberto M. Cesar University of São Paulo, Brazil

Badong Chen Xi’an Jiaotong University, China

Jian Cheng Institute of Automation of CAS, China

Ananda Chowdhury Jadavpur University, India

Donatello Conte Université Francois-Rabelais, Tours, France

Guillaume Damiand LIRIS laboratory, France

Francisco Escolano Universidad de Alicante, Spain

Volkmar Frinken Kyushu University, Japan

Pasquale Foggia University of Salerno, Italy

Edwin R. Hancock University of York, UK

Qiang Ji Rensselaer Polytechnic Institute, USA

Xiaoyi Jiang University of Münster, Germany

Walter G. Kropatsch Vienna University of Technology, Austria

Cheng-Lin Liu Institute of Automation of CAS, China

Zhiyong Liu Institute of Automation of CAS, China

Josep Lladós Universitat Autònoma de Barcelona, Spain

Bin Luo Anhui University, China

Jean-Marc Ogier Université de La Rochelle, France

Shinichiro Omachi Tohoku University, Japan

Marcello Pelillo University of Venice, Italy

Peng Ren China University of Petroleum, China

Kaspar Riesen University of Applied Sciences and Arts, Switzerland

3

Francesc Serratosa Universitat Rovira i Virgili, Spain

Ali Shokoufandeh Drexel University, USA

Andrea Torsello Ca’Foscari University of Venice, Italy

Seiichi Uchida Kyushu University, Japan

Mario Vento University of Salerno, Italy

Richard Wilson University of York, UK

Shuicheng Yan National University of Singapore, Singapore

Changshui Zhang Tsinghua University, China

Local Organizing Committee:

Jian Cheng Institute of Automation of CAS, China

Yan-Ming Zhang Institute of Automation of CAS, China

Ming Li Institute of Automation of CAS, China

Referees

Xiang Bai Xiao Bai Luc Brun

Roberto M. Cesar Badong Chen Jian Cheng

Ananda Chowdhury Donatello Conte Guillaume Damiand

Francisco Escolano Pasquale Foggia Volkmar Frinken

Edwin R. Hancock Qiang Ji Xiaoyi Jiang

Josep Llados Cheng-Lin Liu Zhi-Yong Liu

Bin Luo Jean-Marc Ogier Shinichiro Omachi

Marcello Pelillo Peng Ren Kaspar Riesen

Francesc Serratosa Ali Shokoufandeh Andrea Torsello

Mario Vento Seiichi Uchida Richard Wilson

Shuicheng Yan Changshui Zhang

Additional Referees

Li Chuan Andreas Fischer Llus-Pere De Las Heras

Yinlin Li Xiaodan Liang Muhammad Muzzamil

Luqman Tomo Miyazaki Alessia Saggese

XinggangWang Xu Yang Suiwu Zheng

4

Program at a Glance

May 13

(Wednesday)

8:00-9:00 Registration

9:00-9:20 Opening

9:20-10:20 Keynote Speech 1

10:20-10:50 Tea Break

10:50-12:30 Graph Matching 1

12:30-14:00 Lunch

14:00-15:40 Graph-based Representation 1

15:40-16:10 Tea Break

16:10-17:50 Graph Clustering and Classification 1

May 14

(Thursday)

9:00-10:00 Keynote Speech 2

10:00-10:30 Tea Break

10:30-12:35 Graph-based Applications 1

12:35-14:00 Lunch

14:00-15:40 Graph Matching 2

15:40-16:10 Tea Break

16:10-17:25 Graph Clustering and Classification 2

17:25-18:10 TC15 Session

19:30-22:00 Banquet

May 15

(Friday)

8:40-10:20 Graph Matching 3

10:20-10:50 Tea Break

10:50-12:30 Graph-based Representation 2

12:30-14:00 Lunch

14:00-15:40 Graph-based Applications 2

15:40-16:00 Closing

5

Keynote Speech 1

Revealing Structure in Large Graphs: Szemerédi's Regularity

Lemma and Its Use in Pattern Recognition

Abstract:  Introduced in the mid-1970's as an intermediate step in proving a long-standing

conjecture on arithmetic progressions, Szemeredi's regularity lemma has emerged over time as a

fundamental tool in different branches of discrete mathematics and theoretical computer science.

Roughly, it states that every graph can be approximated by the union of a small number of

random-like bipartite graphs called regular pairs. In other words, the result provides us a way to

obtain a good description of a large graph using a small amount of data, and can be regarded as a

manifestation of the all-pervading dichotomy between structure and randomness. In this talk, I

will provide an overview of the regularity lemma and variations thereof and will discuss its

relevance in the context of structural pattern recognition. 

 

Biography: Marcello Pellilo is Full Professor of Computer Science

at Ca’ Foscari University in Venice, Italy, where he leads the

Computer Vision and Pattern Recognition group. He held visiting

research positions at Yale University, McGill University, the

University of Vienna, York University (UK), the University College

London, and the National ICT Australia (NICTA). He has published

more than 150 technical papers in refereed journals, handbooks, and

conference proceedings in the areas of pattern recognition, computer

vision and machine learning. He is General Chair for ICCV 2017 and

has served as Program Chair for several conferences and workshops

(EMMCVPR, SIMBAD, S+SSPR, etc.). He serves (has served) on the Editorial Boards of the

journals IEEE Transactions on Pattern Analysis and Machine Intelligence (PAMI), Pattern

Recognition, IET Computer Vision, Frontiers in Computer Image Analysis, Brain Informatics, and

serves on the Advisory Board of the International Journal of Machine Learning and Cybernetics.

Prof. Pelillo is a Fellow of the IEEE and a Fellow of the IAPR. His Erdös number is 2.

6

Keynote Speech 2

Understanding Users by Connecting Large Scale Social Graphs

Abstract: With the rapid development of positioning, sensor and smart device technologies, large

quantities of human behavioral data are now readily available. They reflect various aspects of

human mobility and activities in the physical world. The availability of this data presents an

unprecedented opportunity to gain a more in depth understanding of users and provide them with

personalized online experience while respecting their privacy. In this talk, I will describe a vision

toward building a unified user knowledge graph supplemented with knowledge generated from

both online and offline user data, such as social posts, check-in trajectories, and mobile

communication histories. We proposed a graph node similarity measurement in consideration with

both graph structure and descriptive information, and a graph linking/de-anonymization algorithm

based on the measurement. The algorithm has been evaluated in social graphs with thousands of

nodes from Microsoft Academic Search, LiveJournal, and the Enron email dataset, and a social

graph with millions of nodes from Tencent Weibo. Then I will present LifeSpec, a computational

framework for exploring and hierarchically categorizing urban lifestyles, based on large scale

publicly available heterogeneous behavioral data from multiple social networks.

Biography: Xing Xie is currently a senior researcher in Microsoft

Research Asia, and a guest Ph.D. advisor for the University of Science

and Technology of China. He received his B.S. and Ph.D. degrees in

Computer Science from the University of Science and Technology of

China in 1996 and 2001, respectively. He joined Microsoft Research

Asia in July 2001, working on spatial data mining, location based

services, social networks and ubiquitous computing. During the past

years, he has published over 140 referred journal and conference

papers. He has more than 50 patents filed or granted. He currently

serves on the editorial boards of ACM Transactions on Intelligent

Systems and Technology, Springer GeoInformatica, Elsevier Pervasive and Mobile Computing,

Journal of Location Based Services, and Communications of the China Computer Federation. He

has worked as a guest editor of IEEE Transactions on Multimedia and IEEE Intelligent Systems.

In recent years, he was involved in the program or organizing committees of over 70 conferences

and workshops. Especially, he initiated the LBSN workshop series and served as program co-chair

of ACM UbiComp 2011 and program chair of the 8th Chinese Pervasive Computing Conference.

In Oct. 2009, he founded the SIGSPATIAL China chapter which was the first regional chapter of

ACM SIGSPATIAL. He is a member of Joint Steering Committee of the UbiComp and Pervasive

Conference Series. He is a senior member of ACM, the IEEE, and China Computer Federation.

7

Technical Program

May 13

(Wednesday) Session, Title, Author Page # in

Proceedings

8:00-9:00 Registration

9:00-9:20 Opening

9:20-10:20 Keynote Speech 1, Chair: Bin Luo

Revealing Structure in Large Graphs: Szemerédi's Regularity

Lemma and Its Use in Pattern Recognition

Marcello Pelillo (University of Venice, Italy)

10:20-10:50 Break, Group Photo

10:50-12:30 Graph Matching 1, Chair: Andreas Fisher

A First Step Towards Exact Graph Edit Distance Using

Bipartite Graph Matching

Miquel Ferrer, Francesc Serratosa and Kaspar Riesen

77

Consensus of Two Graph Correspondences through a

Generalisation of the Bipartite Graph Matching Algorithm

Carlos Francisco Moreno-García, Francesc Serratosa and

Xavier Cortés

87

Revisiting Volegnant-Jonker for Approximating Graph Edit

Distance

William Jones, Aziem Chawdhary and Andy King

98

A Hypergraph Matching Framework for Refining

Multi-source Feature Correspondences

He Zhang, Bin Du, Yanjiang Wang and Peng Ren

108

12:30-14:00 Lunch

14:00-15:40 Graph-based Representation 1, Chair: Nean-Marc Ogier

Approximation of Graph Edit Distance in Quadratic Time

Kaspar Riesen, Miquel Ferrer, Andreas Fischer and Horst

Bunke

3

Data Graph Formulation as the Minimum-Weight

Maximum-Entropy Problem

Samuel de Sousa and Walter Kropatsch

13

Thermodynamics of Time Evolving Networks

Cheng Ye, Andrea Torsello, Richard C. Wilson and Edwin R.

Hancock

315

A Subpath Kernel for Learning Hierarchical Image

Representations

Yanwei Cui, Laetitia Chapel and Sébastien Lefèvre

34

15:40-16:10 Break

8

16:10-17:50 Graph Clustering and Classification 1, Chair: Peng Ren

On the Influence of Node Centralities on Graph Edit Distance

for Graph Classification

Xavier Cortés, Francesc Serratosa and Carlos F.

Moreno-García

231

A Mixed Weisfeiler-Lehman Graph Kernel

Lixiang Xu, Jin Xie, Xiaofeng Wang and Bin Luo

242

A Quantum Jensen-Shannon Graph Kernel using

Discrete-time Quantum Walks

Lu Bai, Luca Rossi, Peng Ren, Zhihong Zhang and Edwin R.

Hancock

252

Density Based Cluster Extension and Dominant Sets

Clustering

Jian Hou, Chunshi Sha, Xu E, Qi Xia and Naiming Qi

262

May 14 (Thursday)

9:00-10:00 Keynote Speech 2, Chair: Walter Kropatsch

Understanding Users by Connecting Large Scale Social

Graphs

Xing Xie (Microsoft Research Asia, China)

10:00-10:30 Break

10:30-12:35 Graph-based Applications 1, Chair: Francesc Serratosa

Report on the First Contest on Graph Matching Algorithms for

Pattern Search in Biological Databases

Vincenzo Carletti, Pasquale Foggia, Mario Vento and Xiaoyi

Jiang

178

Kite Recognition by means of Graph Matching

Kamel Madi, Hamida Seba, Hamamache Kheddouci,

Charles-Edmond Bichot, Olivier Barge, Christine Chataigner,

Remy Crassard, Emmanuelle Regagnon and Emmanuelle Vila

118

From bags to graphs of stereo subgraphs in order to predict

molecule's properties

Pierre-Anthony Grenier, Luc Brun and Didier Villemin

305

An Entropic Edge Assortativity Measure

Cheng Ye, Richard C. Wilson and Edwin R. Hancock

23

Graph Based Lymphatic Vessel Wall Localisation and

Tracking

Ehab Essa, Xianghua Xie and Jonathan Jones

345

12:30-14:00 Lunch

14:00-15:40 Graph Matching 2, Chair: Luc Brun

9

GEM++: a tool for solving substitution-tolerant subgraph

isomorphism

Julien Lerouge, Pierre Le Bodic, Pierre Héroux and Sébastien

Adam

128

A Graph Database Repository and Performance Evaluation

Metrics for Graph Edit Distance

Zeina Abu-Aisheh, Romain Raveaux and Jean-Yves Ramel

138

Improving Hausdorff Edit Distance Using Structural Node

Context

Andreas Fischer, Seiichi Uchida, Volkmar Frinken, Kaspar

Riesen and Horst Bunke

148

VF2 Plus: An Improved Version of VF2 for Biological Graphs

Vincenzo Carletti, Pasquale Foggia, and Mario Vento

168

15:40-16:10 Break

16:10-17:25 Graph Clustering and Classification 2, Chair: Jian Cheng

Salient Object Segmentation from Stereoscopic Images

Xingxing Fan, Zhi Liu and Linwei Ye

272

Causal Video Segmentation using Superseeds and Graph

Matching

Vijay N. Gangapure, Susmit Nanda, Ananda S. Chowdhury

and Xiaoyi Jiang

282

Fast Minimum Spanning Tree based Clustering Algorithms on

Local Neighborhood Graph

R.Jothi, Sraban Kumar Mohanty and Aparajita Ojha

292

17:25-18:10 TC15 session, Chair: Xiaoyi Jiang

19:30-22:00 Banquet

May 15 (Friday)

8:40-10:20 Graph Matching 3, Chair: Pasquale Foggia

Approximate Graph Edit Distance Computation Combining

Bipartite Matching and Exact Neighborhood Substructure

Distance

Vincenzo Carletti, Benoit Gaüzère, Luc Brun and Mario Vento

188

Multi-layer Tree Matching Using HSTs

Yusuf Osmanlioglu and Ali Shokoufandeh

198

Large-scale Graph Indexing using Binary Embeddings of

Node Contexts

Pau Riba, Josep Llados, Alicia Fornes and Anjan Dutta

208

Attributed Relational Graph Matching with Sparse Relaxation

and Bistochastic Normalization

Bo Jiang, Jin Tang and Bin Luo

218

10

10:20-10:50 Break

10:50-12:30 Graph-based Representation 2, Chair: Xiao Bai

Coupled-Feature Hypergraph Representation for Feature

Selection

Zhihong Zhang, Jianbing Xiahou, Lu Bai and Edwin R.

Hancock

44

Reeb Graphs Through Local Binary Patterns

Ines Janusch and Walter Kropatsch

54

Incremental embedding within a dissimilarity-based

framework

Rachid Hafiane, Luc Brun and Salvatore Tabbone

64

Learning Graph Model for Different Dimensions Image

Matching

Haoyi Zhou, Xiao Bai, Jun Zhou, Haichuan Yang and Yun Liu

158

12:30-14:00 Lunch

14:00-15:40 Graph-based Applications 2, Chair: Josep Llados

Skeletal Graphs from Schrodinger Magnitude and Phase

Francisco Escolano, Edwin R. Hancock and Miguel Angel

Lozano

335

Isometric Mapping Hashing

Yanzhen Liu, Xiao Bai, Haichuan Yang, Jun Zhou and

Zhihong Zhang

325

A Comic Retrieval System Based on Multilayer Graph

Representation and Graph Mining

Thanh-Nam Le, Muhammad Muzzamil

Luqman,Jean-Christophe Burie, and Jean-Marc Ogier

355

Learning High-Order Structures for Texture Retrieval

Ni Liu, Georgy Gimelfarb and Patrice Delmas

365

15:40-16:00 Closing

11

Workshop Venue

The meeting room is on the 3rd floor of Intelligence Building (map below), Meeting Room #1

(智能化大厦 3 层第一会议室). You should enter the campus of CASIA through the north

gate of Automation Building (自动化大厦), and then enter the Intelligence Building

Intelligence Building

12

Notes

13

Notes

14

Notes