ACM Transactions on Multimedia Computing, Communications and Applications (TOMM)

CALL FOR PAPERS

Special Issue on Advanced Approaches for Multiple Instance Learning in Multimedia Applications (AMIL)

Multiple instance learning (MIL) is a form of weakly supervised learning where training instances are arranged in sets, called bags, while the points inside the bags are named instances and a label is provided for an entire bag. The main distinctiveness of the MIL paradigm is that only the labels of the bags are known, whereas the labels of the instances are unknown. It is not similar to classical supervised classification approaches, where the label of each data point is indeed known. This paradigm is gaining interest because it naturally fits various problems and allows leveraging weakly labeled data.

The MIL paradigm has been widely applied in multimedia applications, such as image processing, video processing, signal processing, text processing, and drug design. However, learning from bags raises some unique challenges such as the composition of the bags, the ambiguity of instance labels, and the tasks to be performed.

The objective of this special issue is to provide a forum for researchers to share their recent progresses on multiple instance learning for any type of multimedia data or applications. Papers could cover broad aspects from theoretical to engineering perspectives, including MIL techniques for modeling, MIL algorithms for image, signal, and text processing, and novel MIL designs for computational imaging in various spectral regimes (such as optical, ultrasound, microwave regimes), and so on. Contributions are also welcome concerning applications using MIL from fundamental science to applied research.

Potential topics include, but are not limited to

· MIL architectures for image, video, document and sound processing

· MIL approaches for face detection, pose estimation, object detection/segmentation/tracking, human behavior analysis, and image retrieval

· MIL integrated framework for learning deep representation

· MIL for supervised/unsupervised small object detection

· MIL and hybrid models for real-time computational methods

· New objective functions of MIL for image processing

· Applications for natural, medical, remote sensing research

· MIL and deep learning approaches for multimedia data

Important dates

· Paper submission due: April 30, 2020

· First notification: June 30, 2020

· Revision submission due: August 31, 2020

· Final decision: September 30, 2020

Paper submission and review process

Review papers and the papers outside the areas listed above but related to the overall scope of the special issue are also welcome. Prospective authors can contact the Guest Editors to ascertain interest on such topics. Submission of a paper to ACM TOMM is permitted only if the paper has not been submitted, accepted, published, or copyrighted in another journal. For submission information, please refer to the ACM TOMM journal guidelines (see https://tomm.acm.org/authors.cfm). Manuscripts should be submitted through the online system (https://mc.manuscriptcentral.com/tomm).

Guest editors will make an initial determination of the suitability and scope of all submissions. The review process will be done by following the standard review process of TOMM. Each paper will be reviewed by at least three experts in the field. In general, only two reviewing rounds will be possible, out of which major revision is possible only for the first round. Papers that after the 2nd reviewing round still need major revision will be rejected.

Guest editors

Dr. Pourya Shamsolmoali

Department of Automation, Shanghai Jiao Tong University, Shanghai, China.

E-mail: pshams@sjtu.edu.cn

Prof. Ruili Wang

School of Natural and Computational Sciences, Massey University, Auckland, New Zealand.

E- mail: ruili.wang@massey.ac.nz

Prof. Abdul Hamid Sadka

Brunel Digital Science & Technology Hub and Centre for Media Communications Research, Brunel University, London, UK.

E-mail: abdul.sadka@brunel.ac.uk