Towards Coordinated Bandwidth Adaptations for Hundred-Scale 3D Tele-Immersive Systems

TL;DR

This paper presents TELEVIS, a HVS-aware bandwidth adaptation framework for 3D tele-immersive systems, enabling efficient delivery of large-scale 3D streams over bandwidth-limited networks, improving quality per bandwidth unit.

Contribution

It introduces a novel HVS-compliant adaptive bandwidth management framework and a simulation tool for 3D tele-immersive systems, addressing large data transmission challenges.

Findings

Improved quality per bandwidth in 3D tele-immersive streaming.

Effective HVS-aware adaptation enhances stream delivery.

TELEVIS demonstrates realistic simulation of adaptive streaming.

Abstract

3D tele-immersion improves the state of collaboration among geographically distributed participants. Unlike the traditional 2D videos, a 3D tele-immersive system employs multiple 3D cameras based in each physical site to cover a much larger field of view, generating a very large amount of stream data. One of the major challenges is how to efficiently transmit these bulky 3D streaming data to bandwidth-constrained sites. In this paper, we study an adaptive Human Visual System (HVS) -compliant bandwidth management framework for efficient delivery of hundred-scale streams produced from distributed 3D tele-immersive sites to a receiver site with limited bandwidth budget. Our adaptation framework exploits the semantics link of HVS with multiple 3D streams in the 3D tele-immersive environment. We developed TELEVIS, a visual simulation tool to showcase a HVS-aware tele-immersive system for realistic cases. Our evaluation results show that the proposed adaptation can improve the total quality per unit of bandwidth used to deliver streams in 3D tele-immersive systems.