Special Issue "HPC with Artificial Intelligence based Deep Video Data Analytics: Models, Applications and Approaches"

Submission Deadline: 30 November 2021 (closed)
Guest Editors
Dr. Seungmin Rho, Sejong University, Korea
Dr. S. Vimal, National Engineering College, India
Prof. Naveen Chilamkurti, La Trobe University, Australia


Significance & Novelty:

Video analytics is an important area of research in modern high performance computing paradigm. The modern era has expansion of video data used for modern surveillance and personal data captures. The processing of large video data is indeed a big task. The deep learning based video data analytics is a major platform where most of the researchers focus on the big visual data to modern real time applications. The video data is assumed to be of holding a large spatial and temporal analysis which can be addressed easily with Deep learning to provide the clear pixel level labels with the AI based Deep video data analytics approaches.  Besides, deep learning is an approach to solve the supervised and unsupervised learning problems to address various issues arising due to GPU clusters.


High Performance Computing (HPC) in association with Artificial Intelligence based deep learning is often termed as deep Intelligent HPC, it drives a major shift in the paradigm with data analytics and subsequent data processing. The information in the data centre needs a highly securable and performance viable data processing in a highly secured environment. The today era in academia and industry perspectives need an intelligent HPC infrastructure to analyse, process and validate the data. AI is a good technique to support the various perspectives with a wide range of capability from analysis to storage with good retrieval. The traditional infrastructures in data centres concentrate on the fast retrieval mechanism, but AI based HPC enables a supercomputing mechanism and flexible access with the support of various machine learning and deep learning algorithms. In this special issue, we attempt to assemble recent advances in the deep learning based video analysis and related extended applications.


Objectives and scope:

• To identify Artificial Intelligence techniques in data analytics and computing environment that are suitable for video applications.

• To recognize a wide variety of learning algorithms and how to apply a variety of those algorithms to data.

• To have a good understanding of the fundamental issues and challenges of AI based deep learning: data, model selection, model complexity, etc…

• Integration of heterogeneous computing and big data analytics as a powerful new paradigm to implement the concept of high performance computing in video analytics.

• To introduce the advancements in the computing field to effectively handle and make inferences from voluminous and heterogeneous data.

• State-of-the-art AI approaches need to be improved in terms of data integration, interpretability, security and temporal modeling to be effectively applied to the video data.


Topics of interest include, but are not limited to, the following scope:

In this special issue, we aim to provide a forum for researchers with an interest in efficiency to examine challenging research questions, showcase the state-of-the-art, and share breakthroughs.

• Learning data representation from video based on supervised/unsupervised/semi-supervised learning

• Object detection and recognition

• Action recognition

• Web video understanding using deep learning techniques, including classification, annotation, event detection and recognition, authoring and editing

• Video highlights, summary and storyboard generation

• Segmentation and tracking

• Data collections, benchmarking, and performance evaluation

• Human behavior analysis in real-time surveillance video surveillance using deep learning

• Mathematical foundations of AI in deep learning

AI, Deep learning, Video Data Analytics, HPC, Intelligent Data Processing

Published Papers
  • Image Translation Method for Game Character Sprite Drawing
  • Abstract Two-dimensional (2D) character animation is one of the most important visual elements on which users’ interest is focused in the game field. However, 2D character animation works in the game field are mostly performed manually in two dimensions, thus generating high production costs. This study proposes a generative adversarial network based production tool that can easily and quickly generate the sprite images of 2D characters. First, we proposed a methodology to create a synthetic dataset for training using images from the real world in the game resource production field where machine learning datasets are insufficient. In addition, we have enabled… More
  •   Views:942       Downloads:522        Download PDF