BROWSE
Detail
Conference Paper
A Hybrid XR Framework for Collaborative Scientific Visualization: Rendering-as-a-Service with ParaView and Real-Time Engines
- Authors
- Jin Sung Choi, Yongwan Kim, Suran Park, Ki-Hong Kim
- Issue Date
- 2025-12
- Citation
- ACM SIGGRAPH Asia (SA) 2025, pp.1-2
- Publisher
- ACM
- Language
- English
- Type
- Conference Paper
- Abstract
- We present a hybrid Rendering-as-a-Service (RaaS) architecture for low-latency, multi-user collaboration on TB–PB-scale scientific datasets across heterogeneous devices. It couples a server-side visualization engine (ParaView/VTK) with extended reality (XR) clients (Unity/Unreal) via a collaboration server. The design follows three principles: clear server–client role partitioning (server: high-cost filtering/rendering; clients: input/feedback), authoritative server-driven synchronization, and lightweight intent-only control. In a LAN prototype with one PC and two HMD users, per-request server time stayed under 20 ms; 20 Hz broadcasts maintained coherence of camera pose, selection, and clipping, sustaining interactive frame rates for tens-of-GB datasets. Though full user studies are beyond scope, results show the hybrid design avoids event-handling stalls common in scientific visualization and enables equivalent collaboration across heterogeneous environments.