A training-free pipeline prunes object-centric 3D Gaussian splats by local competition and packs them into deterministic codec-ready atlases, cutting preparation time and payload with modest quality loss.
Traffic Characteristics of Extended Reality
1 Pith paper cite this work, alongside 5 external citations. Polarity classification is still indexing.
abstract
This tutorial paper analyzes the traffic characteristics of immersive experiences with extended reality (XR) technologies, including Augmented reality (AR), virtual reality (VR), and mixed reality (MR). The current trend in XR applications is to offload the computation and rendering to an external server and use wireless communications between the XR head-mounted display (HMD) and the access points. This paradigm becomes essential owing to (1) its high flexibility (in terms of user mobility) compared to remote rendering through a wired connection, and (2) the high computing power available on the server compared to local rendering (on HMD). The requirements to facilitate a pleasant XR experience are analyzed in three aspects: capacity (throughput), latency, and reliability. For capacity, two VR experiences are analyzed: a human eye-like experience and an experience with the Oculus Quest 2 HMD. For latency, the key components of the motion-to-photon (MTP) delay are discussed. For reliability, the maximum packet loss rate (or the minimum packet delivery rate) is studied for different XR scenarios. Specifically, the paper reviews optimization techniques that were proposed to reduce the latency, conserve the bandwidth, extend the scalability, and/or increase the reliability to satisfy the stringent requirements of the emerging XR applications.
fields
cs.GR 1years
2026 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
AtlasLC: Fast Codec-Ready Compression of Object-Centric 3D Gaussian Splatting
A training-free pipeline prunes object-centric 3D Gaussian splats by local competition and packs them into deterministic codec-ready atlases, cutting preparation time and payload with modest quality loss.