Conference Awards
Online Proceedings is now available
PCM 2015 in Korea
PCM 2014 Photos


Important Dates

Special session proposal due:
30-April-2014
Workshop proposal due:
30-April-2014 31 May 2014
Tutorial Proposal Submission:
30-April-2014 31 May 2014
Full/short paper submission:
31-May-2014 11-July-2014
Demo paper submission:
15-June-2014 11-July-2014
Notification of acceptance:
10-Aug-2014 1-Sept-2014
Camera ready submission:
31-Aug-2014 18-Sept-2014



INTERNATIONAL WORKSHOP ON MULTIMEDIA BIG DATA ANALYTICS (MBDA 2014)

IN CONJUNCTION WITH PACIFIC-RIM CONFERENCE ON MULTIMEDIA (PCM), DEC. 1-4, 2014,

Organizers

Zheng-Jun Zha, Hefei Institute of Intelligent Machines, Chinese Academy of Sciences, China
Guo-Jun Qi, IBM T. J. Watson Research Center, USA
Vasileios Mezaris, Information Technologies Institute, Centre of Research and Technology Hellas (CERTH), Greece

Synopsis

“Big Data” has become a ubiquitous term in recent years and multimedia is becoming the “biggest Big Data.” Multimedia big data comes from sources as varied as security cameras, medical imaging, and individuals sharing media on social networks. Cities are installing hundreds of millions of video cameras worldwide for safety, security and law enforcement. Medical institutions are producing huge quantities of diagnostic images. Web users are posting 300 hundred million photos to Facebook per day and uploading 72 video-hours to YouTube per minute. Multimedia data is not just big in volume, but also multi-modal and unstructured. It is the most important and valuable source for insights and information

The availability of massive and complex multimedia data presents a growing need for effective and efficient analytics of the data. This is spurring on tremendous amount of research and development of techniques and platforms for analyzing multimedia big data. This workshop aims to bring together researchers and industrial practitioners interested in multimedia big data analytics. The workshop will provide a venue for the participants to discuss key research issues on multimedia big data. It will collect and seek the recent important research works in this area, summarize the available resources, and exploit potential challenges and possible advanced solutions in terms of theory and practice.

Accepted Papers

Food Image Recognition using Deep Convolutional Features Pre-Trained with Food-related Categories
Kawano Yoshiyuki and Keiji Yanai
Estimating Sentiment Polarity of Web Images Based on User-Generated Tags and SentiWordNet
Katsurai Marie
An Analysis on Visual Recognizability of Onomatopoeia Using Web Images and DCNN features
Shimoda Wataru and Yanai Keiji
Generic Object Categorization using Specific Object Recognition Methods with a Large Number of Web Images
Akiyama Mizuki, Kawano Yoshiyuki, and Yanai Keiji