Call for Paper - Special Issue on Intelligent Multimedia Search
1 Scope and Motivation
Multimedia search plays the core role in almost everywhere in the modern society, including security, industry, education and business. As the increasing number of implemented multimedia sensors, the boosted amount of available multimedia data requires intelligent processing for both efficient and qualitative searching.
Unfortunately, conventional methods for multimedia search are usually deficient and incomplete when dealing with the large-scale data. The unexpected performance mainly comes from the redundant data representation, insufficient inference ability and costly retrieval. Firstly, current multimedia data are much more informative compared with the traditional one, which improves the requirement for the compact data representation; the exiting human-designed methods for multimedia search, however, usually sacrifice their representation compactness for higher searching accuracy, which leads to the redundant representation space and large computation cost. Secondly, in the multimedia searching scenario, each of the single data are not isolated, and inferencing the association among them is important for conducting completely searches; the manually set inference manners in exiting methods, which fails to reveal the data associations generally, thus results in defect searching results. Thirdly, the conventional methods follow the fixed retrieval strategy in the searching process, which leads to inflexible and non-adaptive processing for the unseen data, thus increasing the computational cost.
As a result, intelligent solutions for multimedia search are urgently required. For example, rep- representation learning techniques for compact data representation extraction, thus reducing the size of data space and improving the searching efficiency; graph learning techniques for obtaining the global understanding of the available data, thus producing the complete searching results. Other techniques, such as multimedia data processing, adaptive data structuring and intelligent information retrieval are also important for the efficient and qualitative searching process.
In this special issue, we therefore propose the fundamental theme on Intelligent Multimedia Search, which aims to attract the research focus on efficient and qualitative searching solutions from all research communities, and bring development in the combination of intelligent design and multimedia search areas.
2 Topics
The goal of this special issue is to seek for high-quality and original papers that strive to provide solutions for efficient and qualitative multimedia search, with attention on intelligent techniques. The topics of interest include, but not limited to the following topics:
-
Representation Learning
-
Graph Learning
-
Multimedia Data Processing
-
Adaptive Data Structuring
-
Qualitative Searching Process
-
Vision and Language
-
Image Understanding and Object Recognition
-
Search Algorithms
-
Model Compression and acceleration
-
Heterogeneous classification
-
Semi/self-supervised Learning
-
Multimodal retrieval
-
Recommender System
-
Reinforcement Learning
3 Submission
This special issue “Intelligent Multimedia Search” welcomes submissions of high-quality novel research results. All submissions will be anonymously peer reviewed and will be evaluated on the basis of their technical content.
Papers should be submitted online through the manuscript submission system of the Journal of China Universities of Posts and telecommunications. Remember to select “Special Issue on Intelligent Multimedia Search” as your manuscript type when submitting; otherwise, it might be considered as a regular paper. Information and guidelines on preparation of manuscript are available on the journal website: https://jcupt.bupt.edu.cn.
4 Important Dates
Submission Deadline – February 28, 2022
Final Decision - April 15, 2022
Final Manuscript Submission - April 30, 2022
Publication Date - June 30, 2022
5 Team
Guest Editor-in-Chief:
Gao Xinbo（高新波）, Chongqing University of Posts and Telecommunications
Guest Editors:
Ma Zhanyu（马占宇）, Beijing University of Posts and Telecommunications
Li Zechao（李泽超）, Nanjing University of Science and Technology
Kan Meina（阚美娜）, Institute of Computing Technology, Chinese Academy of Sciences
Wang Xinggang（王兴刚）, Huazhong University of Science and Technology
Jian Meng（简萌）, Beijing University of Technology
发布日期： 2021-12-02 浏览：
356
