Submissions due: 25 February 2021
Publication: September/October 2021
IEEE Computer Graphics and Applications (CG&A) plans a September/October 2021 special issue on “Powering Visualization with Deep Learning,” which advances visualization applications using deep-learning techniques. The great success of deep-learning techniques in computer vision, natural language processing, and speech recognition offers new opportunities for data visualization and analytics. We can leverage these technologies not only to recognize visual representations but also to understand analytic tasks using natural language.
Introducing deep-learning techniques into visualization tasks, however, poses challenges. For example, end-to-end deep-learning techniques require a large amount of labelled data to work optimally. Unlike computer-vision applications relying on natural images or videos that can be conveniently collected, high-quality visualization content (i.e., effective visual representation targeting at some tasks) requires experts to generate and validate, which is extremely difficult to produce at scale. In addition, visualization content has clear contours and textures, which is different from natural images. Therefore, debates exist on the applicability of deep-learning techniques in visualization tasks.
For this special issue, we are soliciting papers that describe algorithms, data structures, tools, and systems that use deep learning or facilitate the use of deep learning for visualization tasks. More specifically, we are looking for contributions that demonstrate practical impact of deep learning on (but not limited to) the following topics:
- Datasets for visualization tasks
- Data-driven quality metrics for visualization
- Deep-learning models and training schemes for visualization tasks
- Predictive visual analytics
- The creation and recommendation of visualization content
- Understanding visualization-oriented natural language
- Visualization captioning
- Deep-learning models for interactive visual analytics
- Visualization for deep-learning tasks
There is a strict 8,000-word limit, with tables and figures equivalent to 200 words each. There is a maximum of 20 references for final manuscripts. Authors should be aware that CG&A cannot accept or process papers that exceed this word limit. Articles should be understandable by a broad audience of computer science and engineering professionals, avoiding a focus on theory, mathematics, jargon, and abstract concepts. All manuscripts must be submitted to ScholarOne Manuscripts by the deadline in order to be considered for publication. Submissions are subject to peer review on both technical merit and relevance to CG&A readership. Accepted papers must be well written and understandable, as the level of editing will be a light copyedit. For accepted papers, authors will be required to provide electronic files for each figure according to the following guidelines: for graphs and charts, authors must submit them in their original editable source format (PDF, Visio, Excel, Word, PowerPoint, etc.); for screenshots or photographs, authors must submit high-resolution files (300 dpi or higher at the largest possible dimensions) in JPEG or TIFF formats.
Please direct any pre-submission correspondence to the guest editors at firstname.lastname@example.org.
- Siwei Fu, Zhejiang Lab
- Jian Zhao, University of Waterloo
- Chris Bryan, Arizona State University
- Yingcai Wu, Zhejiang University