Language-oriented Semantic Communication for Image Transmission with Fine-Tuned Diffusion Model

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Wei, Xinfeng, Tong, Haonan, Yang, Nuocheng, Yin, Changchuan
Natura: Preprint
Pubblicazione: 2024
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866909326042464256
author Wei, Xinfeng
Tong, Haonan
Yang, Nuocheng
Yin, Changchuan
author_facet Wei, Xinfeng
Tong, Haonan
Yang, Nuocheng
Yin, Changchuan
contents Ubiquitous image transmission in emerging applications brings huge overheads to limited wireless resources. Since that text has the characteristic of conveying a large amount of information with very little data, the transmission of the descriptive text of an image can reduce the amount of transmitted data. In this context, this paper develops a novel semantic communication framework based on a text-2-image generative model (Gen-SC). In particular, a transmitter converts the input image to textual modality data. Then the text is transmitted through a noisy channel to the receiver. The receiver then uses the received text to generate images. Additionally, to improve the robustness of text transmission over noisy channels, we designed a transformer-based text transmission codec model. Moreover, we obtained a personalized knowledge base by fine-tuning the diffusion model to meet the requirements of task-oriented transmission scenarios. Simulation results show that the proposed framework can achieve high perceptual quality with reducing the transmitted data volume by up to 99% and is robust to wireless channel noise in terms of portrait image transmission.
format Preprint
id arxiv_https___arxiv_org_abs_2409_17104
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Language-oriented Semantic Communication for Image Transmission with Fine-Tuned Diffusion Model
Wei, Xinfeng
Tong, Haonan
Yang, Nuocheng
Yin, Changchuan
Multimedia
Ubiquitous image transmission in emerging applications brings huge overheads to limited wireless resources. Since that text has the characteristic of conveying a large amount of information with very little data, the transmission of the descriptive text of an image can reduce the amount of transmitted data. In this context, this paper develops a novel semantic communication framework based on a text-2-image generative model (Gen-SC). In particular, a transmitter converts the input image to textual modality data. Then the text is transmitted through a noisy channel to the receiver. The receiver then uses the received text to generate images. Additionally, to improve the robustness of text transmission over noisy channels, we designed a transformer-based text transmission codec model. Moreover, we obtained a personalized knowledge base by fine-tuning the diffusion model to meet the requirements of task-oriented transmission scenarios. Simulation results show that the proposed framework can achieve high perceptual quality with reducing the transmitted data volume by up to 99% and is robust to wireless channel noise in terms of portrait image transmission.
title Language-oriented Semantic Communication for Image Transmission with Fine-Tuned Diffusion Model
topic Multimedia
url https://arxiv.org/abs/2409.17104