Generative Semantic Communication: How Generative Models Enhance Semantic Communications
News
10/03 The Special Session has been scheduled for Wed, 17 Apr, 08:30 - 10:30 (UTC +9) Location: Room 205B, Coex, Seoul, Korea!
09/03 The list of accepted papers for the special session is available! Read it below!
07/09 The Special Session Overview Paper is now available on ArXiv! Read it on ArXiv.
About
Semantic communication is poised to become a fundamental aspect of future AI-driven communications. It holds the potential to regenerate images or videos with semantic equivalence to the transmitted ones at the receiving end, without solely relying on retrieving the exact sequence of bits that were transmitted. However, existing solutions are yet to develop the capacity to construct elaborate scenes based on the partial information received. Undoubtedly, there is a pressing need to strike a balance between the effectiveness of generation methods and the complexity of transmitted information, while potentially considering the goal of communication. To this end, deep generative models (DGM) such as diffusion and score-based models have been starting to show great potential in semantic communication frameworks, revealing the ability to generate semantically consistent content at the receiver side. Indeed, DGMs are extremely powerful in solving image and audio inverse problems and in generating multimedia content even from information that has been heavily degraded from the transmission channel. Therefore, due to such abilities, DGMs can significantly enhance next-generation semantic communication frameworks.
The aim of this special session is to bring together leading researchers in the fields of generative modeling and wireless communication so as to provide advances in generative learning methods in semantic communication that can empower science and technology for humankind.
Topics of Interest
Deep generative models: algorithms and applications
Multi-agent reinforcement/generative learning for semantic communications
Distributed generative learning architectures for semantic communications
Efficient/scalable generative models and training algorithms for semantic communications
Diffusion/Score-based models for emerging topics (virtual reality, autonomous driving, etc.)
Diffusion/Score-based models for inverse problems solution
Semantic communications in emerging wireless networks (virtual reality, autonomous driving, etc.)
Semantic sampling and quantization
Semantic compression
Semantic information pursuit for multimodal data
Intelligent and Concise Networks
Paper submission
Five-page papers do not have to be submitted through ICASSP 2024 submission system!
To submit a paper to the special session, follow these steps:
1. Use the following URL: https://www.google.com/url?q=https://cmsworkshops.com/ICASSP2024/Papers/Submission.asp?SessionType%3DSpecial&source=gmail-imap&ust=1692904565000000&usg=AOvVaw2LT1RNWxj6ZO9voDQRIQGr.
2. From the "Special Session Name" menu, select:
20.8: Generative Semantic Communication: How Generative Models Enhance Semantic Communications.
3. Continue to Author Entry.
If you have already submitted your paper through the regular paper submission, please email at: papers@2024.ieeeicassp.org.
All the submissions will go through peer review. More details on paper submission can be found on ICASSP 2024 Paper Submission.
🔴The deadline for paper submission is September 6, 2023.
List of Accepted Papers
DIFFUSION-BASED SPEECH ENHANCEMENT WITH JOINT GENERATIVE AND PREDICTIVE DECODERS
Hao Shi, Kazuki Shimada, Masato Hirano, Takashi Shibuya, Yuichiro Koyama, Zhi Zhong, Shusuke Takahashi, Tatsuya Kawahara, Yuki Mitsufuji
Kyoto University, Japan
Sony Group Corporation, Japan
ENHANCING SEMANTIC COMMUNICATION WITH DEEP GENERATIVE MODELS: AN OVERVIEW
Eleonora Grassucci, Yuki Mitsufuji, Ping Zhang, Danilo Comminiello
Sapienza University of Rome, Italy
Sony Group Corporation, Japan
Beijing University of Posts and Telecommunications, China
LANGUAGE-ORIENTED COMMUNICATION WITH SEMANTIC CODING AND KNOWLEDGE DISTILLATION FOR TEXT-TO-IMAGE GENERATION
Hyelin Nam, Jihong Park, Jinho Choi, Mehdi Bennis, Seong-Lyun Kim
Yonsei University, Korea
Deakin University, Australia
University of Oulu, Finland
DIFFSC: SEMANTIC COMMUNICATION FRAMEWORK WITH ENHANCED DENOISING THROUGH DIFFUSION PROBABILISTIC MODELS
Zeyu Jiang, Xiaohong Liu, Guoxing Yang, Weizhi Li, Aini Li, Guangyu Wang
Beijing University of Posts and Telecommunications, China
University College London, United Kingdom of Great Britain and Northern Ireland
Peng Cheng Laboratory, China
DIFFUSION MODELS FOR AUDIO SEMANTIC COMMUNICATION
Eleonora Grassucci, Christian Marinoni, Andrea Rodriguez, Danilo Comminiello
Sapienza University of Rome, Italy
SG2SC: A GENERATIVE SEMANTIC COMMUNICATION FRAMEWORK FOR SCENE UNDERSTANDING-ORIENTED IMAGE TRANSMISSION
Minxi Yang, Dahua Gao, Feng Xie, Jiaxuan Li, Xiaodan Song, Guangzhou, Guangming Shi
School of Artificial Intelligence, Xidian University, China
Institute of Technology, Xidian University, China
SEMANTIC-PRESERVING IMAGE CODING BASED ON CONDITIONAL DIFFUSION MODELS
Francesco Pezone, Osman Musa, Giuseppe Caire, Sergio Barbarossa
Sapienza University of Rome, Italy
Technische Universität Berlin, Germany
Organizers
Eleonora Grassucci - Sapienza University of Rome - eleonora.grassucci@uniroma1.it
Yuki Mitsufuji - Sony Group Corporation - yuhki.mitsufuji@sony.com
Ping Zhang - Beijing University of Posts and Telecommunications - pzhang@bupt.edu.cn