中国工程院院刊《信息与电子工程前沿（英文）》

 “智能可视分析”专题

征稿通知

2019年7月31日截稿

最近，以深度学习为代表的人工智能技术在计算机视觉、自然语言处理、语音识别等领域取得了突破性进展。深度学习的巨大成功激发了可视化领域“AI for VIS”的研究趋势，推动了若干重要进展。首先，最近的语义嵌入技术和深度神经网络技术带来了复杂数据的新表示方法，在时空数据、事件序列数据可视分析等若干案例中展现了揭示潜在语义的功能。其次，计算机视觉领域中图像内容理解等技术的进步，启发了数据驱动的质量度量等可视化理解的研究。再次，可视化的主要任务是生成数据的高质量可视表达；针对这一任务，AI技术催生了数据驱动的可视化设计方法。此外，还有预测式可视分析等大量可视分析应用受益于AI技术的进步。

但在可视化中应用AI技术仍然存在着理论和实践中的若干难题。首先，训练数据获取困难。计算机视觉中可以相对容易地获取大量有标签的图像数据。但在可视化中，很少有针对专门任务的有标签数据集。其次，自然图像中的视觉对象往往具有稳定的外形轮廓和纹理特征，但信息可视化中的视觉对象的特征与之存在较大差异。当前流行的深度神经网络是否能够用于信息可视化任务还存在争议。再次，当前的AI模型往往需要向量化的表达，而针对特定可视化任务的设计空间的形式化表达仍然是一个需要探索的问题。最后，一些交互可视分析任务本质上就是交互式、渐进式的，如何融合AI技术与交互技术仍然是一个挑战。

针对以上需求，为推动我国在智能可视分析领域的研究工作，为数据分析和可视化提供先进理论和应用成果，中国工程院院刊《信息与电子工程前沿（英文）》（FITEE）邀请潘云鹤院士担任主编，浙江大学CAD&CG国家重点实验室陈为教授担任执行主编，筹备“智能可视分析”专题。为推动国际合作，专题特别鼓励中国学者与国外学者合作投稿。

专题编委会：

主编：

潘云鹤    教授，中国工程院院士，浙江大学

执行主编：

陈为       教授，浙江大学CAD&CG国家重点实验室

编委（按姓氏字母序）：

Steffen Koch      博士，University of Stuttgart，德国

Tobias Schreck  教授，Graze University of Technology，奥地利

Han-Wei Shen   教授，The Ohio State University，美国

夏佳志                副教授，中南大学，中国

Cong Xie            博士，Facebook，美国

Ye Zhao             教授，Kent State University， 美国

联系人：

中南大学，夏佳志，xiajiazhi@csu.edu.cn

FITEE编辑部：翟自洋，jzus_zzy@zju.edu.cn，86-571-88273162

专题学科范畴、文章类型、投稿要求、期刊简介等，请见下方Call for Papers。

欢迎国内外相关领域专家、学者踊跃投稿！

FITEE Special Issue on

AI for Visualization

Call for Papers

Submission Deadline: July 31, 2019

Artificial intelligence (AI) techniques, such as deep learning, have achieved breakthrough in various tasks in recent years, such as computer vision, natural language processing and speech recognition. The great success of AI techniques has inspired a wide range of visualization applications. First, while data transformation is one of the core steps in visualization, recent embedding techniques and deep neural networks provide new representations for complex data, disclosing latent features and enabling efficient operations. Second, inspired by recent advances of understanding visual contents in computer vision, researchers have begun to introduce AI in visualization, such as data-driven quality metrics. Third, while the main goal of visualization is to create new visual representation, data-driven design has become a new methodology of visualization generation. In addition, a great many of applications, such as predictive visual analytics, have also been facilitated by recent AI techniques.

Introducing recent AI techniques into visualization applications, however, yields new methodological and practical challenges that need to be addressed. First, unlike computer vision that can collect training data conveniently, high-quality training data for specific visualization tasks are scare. Second, visual contents in information visualization are rather different from natural visual objects. The latter often have specific contours and textures. Therefore, despite the high performance of deep neural networks in understanding natural images and videos, there are still debates on the applicability of deep neural networks in information visualization. Third, well-formulated design spaces for specific visualization suitable for vectorized representation of deep learning are yet to be investigated. Finally, many tasks of visual analytics are inherently interactive and progressive, integrating user interaction into learning process is also a challenge.

For this special issue, we are looking for submissions that describe algorithms, data representations, tools and systems for visualization tasks based on AI techniques. We also welcome evaluations providing inspiring guidelines on the use of AI techniques in visualization and surveys providing comprehensive discussions on current development of "AI for VIS." More specifically, we are looking for contributions that demonstrate practical impact of AI on (but not limited to) the following topics:

 Data synthesis for training of specific visualization tasks

• Data-driven quality metrics for visualization

• Data-driven design of visualization

• Deep learning model for visualization contents

• Interactive AI techniques for visual analytics

• Predictive visual analytics

To promote international cooperation, we encourage submissions that are co-authored by domestic and international researchers. We also highly recommend the submission of multimedia to accompany each article as it may significantly increase the visibility, downloads and citations.

All submitted manuscripts must be written in English and must not be under consideration elsewhere for publication. The authors must follow the FITEE guidelines (http://www.jzus.zju.edu.cn/manuscript.php) for preparation of their manuscripts. Either Word or LaTeX format is acceptable. When Word is used, please keep the layout of the text as simple as possible, e.g., single column, 1.5 lines spacing, 10.5 pt font size, and Times New Roman font. When LaTeX is used, a template is available at http://www.jzus.zju.edu.cn/download/FITEE_LaTex_template.zip

Templates for accepted papers:

Word: http://www.jzus.zju.edu.cn/download/FITEE_sample.doc

LaTeX: http://www.jzus.zju.edu.cn/download/FITEE_LaTex_template.zip

(At the initial submission stage, authors do not need to use these templates. Only when asked to revise manuscripts after peer reviews, these templates should be used.)

FITEE is an international peer-reviewed journal launched by the Chinese Academy of Engineering (CAE) and Zhejiang University, and co-published by Springer & Zhejiang University Press. FITEE aims to publish the latest implementation of applications, principles, and algorithms in the broad area of Electrical and Electronic Engineering.

All articles published in this special issue will be indexed by SCI-E and will be available from http://www.springer.com/computer/journal/11714, www.jzus.zju.edu.cn, as well as http://engineering.cae.cn/fitee. Please note that all articles will undergo international peer review and Crosscheck processes before acceptance, to ensure that the special issue is of high quality, original, and thought-provoking.

We look forward to your contribution to this special issue. Please send your manuscript via http://www.editorialmanager.com/zusc/. Remember to choose article type “S.I.-AI4VIS”.

Manuscript submission by

July 31, 2019

Acceptance notification by

Nov. 15, 2019

Publication date:

Jan. 31, 2020

Editorial Board:

Editor-in-Chief:

Prof. Yunhe Pan     Academician of CAE, Zhejiang University, China

Executive Lead Editor:

Prof. Wei Chen      CAD&CG State Key Lab, Zhejiang University, China

Editors (in alphabetical order by last name):

Dr. Steffen Koch       University of Stuttgart, Germany

Prof. Tobias Schreck   Graze University of Technology, Austria

Prof. Han-Wei Shen     The Ohio State University, USA

Assoc. Prof. Jiazhi Xia   Central South University, China 

Dr. Cong Xie               Facebook, USA

Prof. Ye Zhao             Kent State University, USA

For inquiries regarding this special issue, please contact:

Jiazhi Xia

Central South University, China

E-mail: xiajiazhi@csu.edu.cn

Editorial Office:

Ziyang Zhai (Managing Editor)

jzus_zzy@zju.edu.cn

86-571-88273162

---------------------

首发于微信公众号“信息与电子工程前沿FITEE”（fitee_cae)。

作者：陈为、夏佳志，等。