Forging a Close Relationship with Multimedia Communities

EEE MultiMedia magazine was founded in 1994 and was the first IEEE publication in the multimedia area. MM serves the community of scholars, developers, practitioners, and students who are interested in multiple media types, used harmoniously together, for creating new experiences. In early 2014, the MM editorial board launched several initiatives to strengthen its collaboration with various multimedia communities in an effort to reach out to a wider range of audience and foster a more efficient publication process. One of the first efforts was to collaborate with the IEEE International Conference on Multimedia & Expo (ICME), the flagship multimedia conference that has been sponsored by four IEEE societies since 2000, to facilitate the publication of extended versions of top ICME papers in MM via a “fast track” review and publication process. In May 2014, the authors of the top 26 ICME 2014 papers were invited to submit extended versions (with at least 30 percent new material) of their papers to this fast track special issue scheduled to be published in the October– December 2014 issue. We received 15 submissions that span various topics such as audio and video coding, vision and pattern analysis, object tracking, quality assessment, and social media. After a rigorous peer-review process, eight of those submissions were accepted for this special issue, now titled “Hot Topics in Multimedia Research.” (Several other fine submissions were also accepted and will be published in early 2015.) A significant subset of the accepted articles address issues in visual analysis and tracking. In “Local Stereo Matching with Improved Matching Cost and Disparity Refinement,” Jianbo Jiao, Ronggang Wang, Wenmin Wang, Shengfu Dong, Zhenyu Wang, and Wen Gao present a technique for improving local stereo matching. They propose a new cost measure to improve the initial matching performance followed by a secondary refinement technique to remove the remaining outliers. In “Multimodal Feature Fusion for 3D Shape Recognition and Retrieval,” Shuhui Bu, Shaoguang Cheng, Zhenbao Liu, and Junwei Han present a deep learning framework to fuse 3D shape and 2D view-based features for 3D shape recognition and retrieval. Gaze-tracking technology is highly valuable in many interactive and diagnostic applications. In “Real-Time Gaze Estimation with Online Calibration,” Li Sun, Mingli Song, Zicheng Liu, and Ming-Ting Sun present a novel 3D-model-based gaze-estimation system using a single consumer depth camera (Kinect) with online calibration to constantly improve person-specific eye parameters. The article “Latent Subspace Projection Pursuit with Online Optimization for Robust Visual Tracking” by Risheng Liu, Wei Jin, Zhixun Su, and Changcheng Zhang proposes an online subspace learning technique to address the problem of feature extraction for visual tracking. In “Online Learning a High-Quality Dictionary and Classifier Jointly for Multitask Object Tracking,” Baojie Fan, Yang Cong, Yingkui Du, Hao Gao, and Yangdong Tang formulate object tracking as a binary classification problem in a particle filter framework. By minimizing an objective function that takes into account both reconstruction and classification errors, they demonstrate how to obtain a high-quality dictionary and optimal linear classifier jointly. Forging a Close Relationship with Multimedia Communities