Home // INTELLI 2024, The Thirteenth International Conference on Intelligent Systems and Applications // View article

Enhancing Light Field Video Compression Efficiency via View Selection and Synthesis Techniques

Authors:
Tala Bazzaza
Maissan Bazazeh
Morteza Rashidinia
Weige Qian
Hamid Reza Tohidypour
Panos Nasiopoulos

Keywords: Light field video compression, prediction structures, view synthesis.

Abstract:
Light field (LF) imaging represents a transformative technology evolving to replicate human like visual data and emulate our visual environment. Departing from traditional single viewpoint cameras, light field cameras capture scenes from multiple perspectives, preserving realistic parallax and capturing the direction of light rays. Despite its potential, the suression by strategically omitting certain views during transmission, and then recreating them at the receiver's end through a specialized synthesis method tailored for light fields. The performance evaluation reveals a delicate balance between bandwidth efficiency and image reconstruction quality in the LF compresbstantial increase in data capture underscores the need for efficient compression techniques. This paper investigates the possibility of enhancing LF data compsion and transmission. We believe that a more efficient view synthesis approach that capitalizes on all directional light field views, holds the promise of enhancing LF compression performance.

Pages: 43 to 46

Copyright: Copyright (c) IARIA, 2024

Publication date: March 10, 2024

Published in: conference

ISSN: 2308-4065

ISBN: 978-1-68558-132-9

Location: Athens, Greece

Dates: from March 10, 2024 to March 14, 2024