CLOSED: Special Issue on Visual Computing with Deep Learning – Call for Papers


Final submissions due: CLOSED
Publication date: 
March/April 2019

The great success of deep learning techniques in computer vision, speech recognition, and natural language processing has recently attracted much attention. While machine learning techniques have long been used to solve a wide range of graphics and visualization problems, most of them rely on problem-specific “feature engineering” to extract favorable features from the training data, which is often a manually-tweaked, time-consuming process and usually does not generalize well. Deep learning techniques, on the other hand, are capable of automatically discovering features appropriate for a specific task from raw data, which reduces the need for feature engineering and makes it easier to develop end-to-end solutions. The recent advances in Generative Adversarial Networks (GAN) and reinforcement learning methods show their potential for data generation and action planning. It is expected that the use of deep learning techniques can significantly advance the performance of many state-of-the-art graphics and visualization algorithms.

Unlike computer vision applications, which mainly focus on visual content analysis and understanding, graphics and visualization tasks must often create visual content (e.g., synthesizing an image, generating an animation sequence, visualizing and interpreting spatial-temporal data) that exhibits the high quality to be used in entertainment or visualization applications. Furthermore, end-to-end deep learning techniques require a large amount of labelled data to work optimally. This raises an additional challenge because, unlike computer vision, which relies on natural images or video that can be conveniently collected on Internet, high-quality synthesized visual content with proper labeling is rare. Finally, different from many vision tasks where automation is the ultimate goal, creating visual content is often an interactive, progressive process. Therefore, user interaction must be integrated into the learning and run-time computation process.

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 graphics and visualization tasks. More specifically, we are looking for contributions that demonstrate practical impact of deep learning on (but not limited to) the following topics:

  • Visual analytics applications
  • Object/scene reconstruction from RGB/RGBD images
  • Shape analysis and synthesis
  • Appearance capture and modeling
  • Global illumination and real-time rendering
  • Sound synthesis and rendering
  • Physics-based simulation of fluids and deformable objects
  • Performance-based face/body animation
  • Computational photography
  • Deep learning models and training schemes for visual content creation

Guest Editors

Please direct any correspondence before submission to the guest editors:

Submission Guidelines

Nondepartment articles submitted to IEEE CG&A should not exceed 8,000 words, including the main text, abstract, keywords, bibliography, biographies, and table text, where a page is approximately 800 words. Articles should include no more than 10 figures or images. Each 1/4 page figure, image, and table counts for approx. 200 words. Note that all tables, images, and illustrations must be appropriately scaled and legible; larger elements should be accounted for accordingly with respect to word count. Please limit the number of references to the most relevant and ensure to delineate your work from relevant past articles in CG&A. Furthermore, avoid an excessive number of references to published work that might only be marginally relevant. Consider instead providing such pertinent background material in sidebars for non-expert readers. Visit the CG&A style and length guidelines at We also strongly encourage you to submit multimedia (videos, podcasts, and so on) to enhance your article. Visit the CG&A supplemental guidelines at

Please submit your paper using the online manuscript submission service at When uploading your paper, select the appropriate special issue title under the category “Manuscript Type.” Also, include complete contact information for all authors. If you have any questions about submitting your article, contact the peer review coordinator at