Abstract With the emergence of the Internet of Things (IoT) and the rise of shared multimedia content on social media networks, available datasets have become increasingly heterogeneous. Several multimodal techniques for detecting events in data of different types and formats have emerged. Those techniques implement various detection algorithms and present different trade-offs in terms of data fusion. Unfortunately, little is known about their underlying detection mechanisms, as existing comparisons are limited to either unimodal event detection techniques or specific types or representations for multimodal techniques. Understanding the behavior of multimodal event detection techniques remains an acute open research problem. In this work, we present a systematic literature review of multimodal event detection techniques. We describe how various techniques leverage information from different modalities through data fusion. We further propose a novel taxonomy of multimodal event detection techniques according to their temporal orientation and the inner workings of their detection mechanism. Finally, we analyze the datasets and metrics used in previous works as well as their reported results. Our survey allows to uncover the properties of each approach and discuss future research directions in this field.

Manuel Mondal

Mourad Khayati

Hông-Ân Sandlin

Philippe Cudré-Mauroux

The VLDB Journal

该杂志致力于发表数据管理领域的学术贡献，例如数据库系统技术和信息系统，包括它们的架构和应用程序。此外，该期刊的范围仅限于该期刊编辑委员会综合专业知识所涵盖的数据管理领域。欢迎提交具有实质性理论成分的投稿，但 VLDB 期刊希望此类投稿也包含系统成分。对于数据挖掘，该期刊将处理系统问题发挥重要作用的提交。我们用于确定数据挖掘论文是否在范围内的因素包括：提交的目标是与数据挖掘相关的系统问题，例如，通过涵盖与数据库引擎或其他数据管理功能的集成。提交的贡献建立在（而不是而不是简单地引用）已经在数据库渠道发表的作品，例如 VLDBJ、ACM TODS、PVLDB、ACM SIGMOD、IEEE ICDE、EDBT。期刊的编辑委员会对提交的主题拥有必要的专业知识。传统的、独立的缺乏上述或类似特征的数据挖掘论文不在本刊范围内。与上述类似的标准适用于来自其他领域的提交，例如信息检索和地理信息系统。

The VLDB Journal is an academic journal dedicated to contributions in the field of data management, including database system technologies and information systems, along with their architectures and applications. The scope of this journal is limited to the areas of data management covered by the expertise of the editorial board. We welcome submissions with substantial theoretical components, but the VLDB Journal also expects these submissions to include system components. For data mining, the journal will consider submissions where system issues play a significant role. Factors we use to determine whether data mining papers fall within scope include whether the submission addresses system issues related to data mining, such as integration with database engines or other data management functionalities. Contributions should build upon (rather than merely cite) existing works published in database channels, such as VLDBJ, ACM TODS, PVLDB, ACM SIGMOD, IEEE ICDE, and EDBT. The editorial board possesses the necessary expertise to evaluate the submitted topics. Traditional, standalone data mining papers lacking the aforementioned or similar characteristics are not within the scope of this journal. Similar standards apply to submissions from other fields, such as information retrieval and geographic information systems.

VLDB Journal 是一本专注于数据管理领域的学术期刊，涵盖数据库系统技术和信息系统，包括其架构和应用程序。该期刊的范围仅限于编辑委员会的专业知识所涵盖的数据管理领域。我们欢迎具有实质性理论成分的投稿，但 VLDB Journal 希望这些投稿也包含系统方面的内容。在数据挖掘方面，期刊将处理那些系统问题发挥重要作用的投稿。我们判断数据挖掘论文是否符合范围的因素包括：提交的目标是否与数据挖掘相关的系统问题相关，例如，是否涉及数据库引擎或其他数据管理功能的集成。投稿的贡献应建立在已有的数据库领域的研究基础上，而不仅仅是简单引用，例如 VLDBJ、ACM TODS、PVLDB、ACM SIGMOD、IEEE ICDE 和 EDBT。期刊的编辑委员会具备必要的专业知识来评估提交的主题。传统的、独立的缺乏上述或类似特征的数据挖掘论文不在本刊的范围内。来自其他领域的提交，例如信息检索和地理信息系统，也适用类似的标准。

Category	Quartile
计算机科学	2
计算机科学, 计算机硬件	2
计算机科学, 计算机信息系统	3

A survey of multimodal event detection based on data fusion

ivySCI AI Smartly Parses PDF, Answers Researchers' Questions, and Helps You Understand Papers in Seconds

Journal Info

Category	Quartile
COMPUTER SCIENCE, INFORMATION SYSTEMS	2