Mining Images and Videos from Social Media
For the first time in human history, literally billions of images are available to anyone with an internet connection. At the same time, people are capturing and sharing millions of images each day, and more than three days-worth of video are shared on average each minute. This deluge of social media contains a great deal of information about people, places, and events, and the challenge is to find the signal amidst a background of noise. The intent of this workshop is to facilitate discussion and interest in methods for using visual processing techniques to mine images from social media to gain insight into the lives of people, detect events, predict popular products, elections, and trends, and more. We invite papers (two-column format, 6 pages, ICME format) that explore new challenges and possibilities from mining images and videos from social media. We are especially interested in applications that aggregate information from large numbers of images and video frames (e.g. tens of thousands or millions of images) to produce a result.
Topics include, but are not limited to:
- real-time 3D modeling of events and places
- event detection from social images and video
- data mining from images and video in social networks
- trend detection and prediction from social media
- recognizing people and places in social media
- social media summarization
- other applications of mining social media
Paper Submission: CMT Paper Submission Site and choose "Mining Images and Videos from Social Media" as the track.
To encourage student participation, a best student paper award will be given.
Workshop Paper Submission: March 14, 2013
Notification of Workshop Paper Acceptance: April 15, 2013
Camera-Ready Paper Due: April 30, 2013
Papers will be reviewed and scored by 3 reviewers (double-blind) to select the accepted papers.