Semantic Communication Based Lossless Distributed Video Coding for Error Prone Channels

Video coding is crucial for modern communications, but traditional methods with complex encoders limit their utility in many emerging consumer electronic applications. Distributed Video Coding (DVC), which shifts encoding complexity to the decoder, is a promising alternative but requires decoder fee...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Proceedings of IEEE International Symposium on Consumer Electronics S. 1 - 4
Hauptverfasser: Samarathunga, Prabhath, Jayasinghe, Udara, Ganearachchi, Yasith, Fernando, Thanuj, Senanayake, Nimesha, Fernando, Anil
Format: Tagungsbericht
Sprache:Englisch
Veröffentlicht: IEEE 11.01.2025
Schlagworte:
Autoencoders
Codecs
Complexity theory
Consumer electronics
Decoding
Deep Neural Networks
Distributed Video Coding
Encoding
Propagation losses
Semantic communication
Signal to noise ratio
Turbo code
Turbo codes
Video coding
ISSN:2158-4001
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:Video coding is crucial for modern communications, but traditional methods with complex encoders limit their utility in many emerging consumer electronic applications. Distributed Video Coding (DVC), which shifts encoding complexity to the decoder, is a promising alternative but requires decoder feedback, limiting its practicality. We propose a semantic communication-based DVC system for lossless video coding, featuring a low-complexity semantic-based video encoder coupled with turbo channel code. The proposed codec significantly outperforms in terms of encoder complexity while maintaining better rate distortion performance at low channel signal-to-noise ratios compared to the state-of-the-art Versatile Video Coding based codec in power constrained consumer video sensors.
ISSN:2158-4001
DOI:10.1109/ICCE63647.2025.10929820