International Workshop on

Multimedia Artworks Analysis (MMArt 2017)

in conjunction with ICME 2017
July 10--14, 2017, Hong Kong


May 31, 2017: Workshop program is announced.

Apr. 24, 2017: Accepted paper list

Mar. 5, 2017: Deadline of paper submission has been extended to Mar. 13. Submission site.

Feb. 20, 2017: Workshop paper submission begins.



In addition to classical art types like paintings, sculpture, music, dance, and movie, new types of artworks emerge as the advancement of social platforms, media capturing devices, and media processing tools. Large volumes of user-generated content or professionally edited content are shared and disseminated on the web. Internet users can browse, comment, edit, and sometimes creatively re-compose this content to generate new media, giving new viewpoints or interesting applications. Novel multimedia artworks, therefore, emerge rapidly in the era of social media and big data. For example, is a social platform where amateur artists/illustrators can share their illustrations/comics, and internet users can annotate, rate, and collect favorite works. The ever increasing amounts of illustrations/comics on this platform give rise to challenges of automatic classification, indexing, and retrieval that have been studied widely in other areas but not this emerging type of artwork.

We encourage theoretical, experimental, and methodological developments to solve the wide range of challenges derived from multimedia artworks in traditional forms like paintings, photos, movies, as well as novel forms like comics, animation, gaming and VR/AR. Topics include, but are not limit to anlaysis and applications of the following areas:


Content synthesis and collaboration; creation of novel artworks; connecting real-world art with digital artworks


Content authoring, composition, summarization, and presentation; multimodality integration

Indexing and retrieval

Novel features and structure to index multimedia artworks; retrieval interface and model; socially-aware analysis


Machine learning for multimedia artworks; classification and pattern recognition for multimedia artworks


Interaction on various devices; user in the loop of computation; human factors in artworks


Dataset development; evaluation of systems for multimedia artworks; design of user study

Novel applications

Novel application scenarios; development of novel challenges and perspectives

Important Dates

Paper submission deadline: 13 March 2017   3 March 2017

Notification of acceptance: 7 April 2017

Camera-ready due: 19 April 2017

Workshop date: 10 or 14 July 2017


Please submit your papers via the CMT submission system,
following the Authors' Author Information and Submission Instructions in the ICME main website.


Date: Monday, July 10, 2017
Room: Salon III

Keynote Session, Session Chair: Wei-Ta Chu, National Chung Cheng University, Taiwan


Keynote Talk: Dr. Ying CAO (City University of Hong Kong)
Talk title: Data-driven Computational Manga: Models, Methods and Applications

In this talk, Dr. Cao will mainly talk about how to leverage manga data to teach computer to understand the nuance of manga
in terms of style, structure and functionality, and build data-driven models and algorithms that can benefit a variety of applications
including manga creation, editing and consumption.

Oral Session 1, Session Chair: Wei-Ta Chu, National Chung Cheng University, Taiwan


Prediction of Users' Facial Attractiveness on an Online Dating Website
Xiaoxue Zang, Toshihiko Yamasaki, Kiyoharu Aizawa, Tetsuhiro Nakamoto,
Eitaro Kuwabara, Shinichi Egami, and Yusuke Fuchida
The University of Tokyo, Japan


Automatic Generation of Sketch-like Pencil Drawing from Image
Jingwei Zhang, Ran-Zan Wang, and Dan Xu
Yunnan University, China


Coffee Break

Oral Session 2, Session Chair: Toshihiko Yamasaki, The University of Tokyo, Japan


Partial Style Transfer Using Weakly Supervised Semantic Segmentation
Shin Matsuo, Wataru Shimoda, and Keiji Yanai
University of Electro-Communications, Japan


Fashion Clothes Matching Scheme Learned from Fashionista's Suggestions in Microblog
Guangyu Gao, Yihang Zhang, and Songyang Du
Beijing Institute of Technology, China


High-level Semantic Photographic Composition Analysis and Understanding with Deep Neural Networks
Min-Tzu Wu, Tse-Yu Pan, Wan-Lun Tsai, Hsu-Chan Kuo, and Min-Chun Hu
National Cheng Kung University, Taiwan


DeepSheet: A Sheet Music Generator Based on Deep Learning
Yu-Lun Hsu, Chi-Po Lin, Bo-Chen Lin, Hsu-Chan Kuo, Wen-Huang Cheng, and Min-Chun Hu
National Cheng Kung University, Academia Sinica, Taiwan

Workshop History

We had the first MMArt workshop at ICME 2016, held in Seattle, USA. We attracted more than 20 researchers from many countries, and had a small but inspiring workshop. One keynote talk was given by Prof. Feng Liu from Portland State University, with talk title "Computational Stereoscopic Photography for Immersive Visual Experience". We had an oral session consisting of four papers, coming from diverse areas, i.e., The Netherlands, Japan, and USA.

Workshop Organizers

Wei-Ta Chu

National Chung Cheng University, Taiwan

Toshihiko Yamasaki

The University of Tokyo, Japan

Takaaki Shiratori

Oculus Research, USA


Technical Committee Members

Gustavo Carneiro

The University of Adelaide, Australia

Ying Cao

City University of Hong Kong, Hong Kong

Naoki Chiba

Rakuten Institute of Technology, Japan


Ahmed Elgammal

Rutgers University, USA

Richang Hong

Hefei University of Technology, China

Bogdan Ionescu

University Politehnica of Bucharest, Romania


Koichi Kise

Osaka Prefecture University, Japan

Shigeru Kuriyama

Toyohashi Institute of Technology, Japan

Yusuke Matsui

National Institute of Informatics, Japan


Akihiko Shirai

Kanagawa Institute of Technology, Japan

Shih-Wei Sun

Taipei National University of the Arts, Taiwan

Mei-Chen Yeh

National Taiwan Normal University, Taiwan


Satoru Fukayama

National Institute of Advanced Industrial Science and Technology, Japan

James McCann

Disney Research Pittsburgh, USA

Eakta Jain

University of Florida, USA