Inference algorithms for semantic knowledge extraction based on deep architectures for context-aware holographic communication

Inference algorithms for semantic knowledge extraction based on deep architectures for context-aware holographic communication

Main Goal

The goal of the project is to develop inference algorithms for semantic knowledge extraction based on deep architectures for context-aware HC by analyzing and combining effective methods and algorithms for 3D human body model acquisition; developing semantic knowledge extraction with deep neural networks to predict human behavior, including analysis of biometric modalities; developing methods for context-aware optimization of network resource allocation for the purpose of creating a multi-party from-capturing-to-rendering HC framework.

Tasks

The main tasks are related to building new theoretical foundations, scientific models, abstractions, and concepts of semantic knowledge extraction from human behavior and biometric modalities with purpose of building the HC framework. The focus will be on:

  • Create a theoretical model of a context-aware HC framework for multisensory communication;
  • Develop improved algorithms for human body modeling with real-time application in 3D space taking into account changes in the multi-view environment;
  • Develop effective deep architectures for semantic knowledge extraction for human activity recognition;
  • Develop semantic inference algorithms and semantic communication strategies to incorporate knowledge representation in communication strategies;
  • Develop novel heterogeneous data compression framework based on semantic knowledge;

Outcome

The outcome of this work is anticipated to have important implications in the area of multisensory communication, a priority area with societal importance. In addition, the results of the research work will contribute to several other research areas including computer vision, machine learning, computer graphics, compression/information theory, networking, high performance computing.

The project is funded by the National Science Fund and is with duration 3 years, starting year 2020. Coordinator of the project is Technical University of Sofia and leader of the project is Assoc. Prof. Agata Hristova Manolova, Ph.D.

Posts related to this project

TeliLab members related to this project

Krasimir Tonchev

Krasimir Tonchev

Senior Researcher
Nikolay Neshov

Nikolay Neshov

Affiliate Researcher
Nicole Christoff

Nicole Christoff

Affiliate Researcher
Agata Manolova

Agata Manolova

Affiliate Researcher

Publications related to this project

3D Human Body Models Compression and Decompression Algorithm Based on Graph Convolutional Networks for Holographic communication

Ivaylo Bozhilov; Krasimir Tonchev; Agata Manolova; Radostina Petkova
2022 25th International Symposium on Wireless Personal Multimedia Communications (WPMC)

Context-Aware Holographic Communication Based on Semantic Knowledge Extraction

Agata Manolova, Krasimir Tonchev, Vladimir Poulkov, Sudhir Dixir, and Peter Lindgren.
Wireless Personal Communications (2021): 1-13.

Research Fields

, , , , , , , , , , , ,