We aim to democratize extended reality (XR) systems research, development, and benchmarking by providing
A reference open source end-to-end XR system testbed, with state-of-the-art components connected with a modular runtime
A benchmarking methodology for XR systems, including reference system configurations, applications, data sets, and metrics
A community where the multidisciplinary XR systems research and development stakeholders come together
The ILLIXR group is excited to share two new papers advancing next-generation XR systems: Boba, accepted to ECCV 2026, makes physics-based Gaussian digital twins practical for interactive XR and scalable simulation; and SemanticXR, accepted to IEEE TVCG / ISMAR 2026, enables real-time, open-vocabulary semantic mapping and querying on low-power XR devices.
Ada: A Distributed, Power-Aware, Real-Time Scene Provider for XR — winner of Best Paper Award at IEEE ISMAR 2025 and selected for publication in IEEE TVCG. The paper introduces the first open, low-latency, low-power, high-fidelity XR scene provisioning system.
Researchers from the City University of New York (CUNY), leveraging the ILLIXR (Illinois Extended Reality) platform in collaboration with the NSF FABRIC testbed, have developed a novel machine-learning framework to detect and defend against stealthy misreporting attacks in software-defined immersive environments.
We are excited to announce a new release of ILLIXR (ILLinois eXtended Reality testbed) v4.1 and our ISMAR/TVCG paper with the first open low power, low latency, and high fidelity scene reconstruction and delivery system for XR. This work is our next milestone in distributed XR – harnessing remote computation via wireless networking – to provide richer XR experiences with low power, comfortable XR devices.
The full tutorial video is available at this link.
The slides for the talk are available.
Tutorial on ILLIXR at ASPLOS on 3/30/25 from 8 am to 11:30 pm CST.
We are very excited to announce a major new release of ILLIXR (ILLinois eXtended Reality testbed) - a significant milestone towards our grand challenge goal of comfortable, mobile, all day, rich immersive experiences. The stringent power budgets of mobile and lightweight XR devices imply that not all immersive computations can be performed locally — remote offloading or Distributed XR is essential. ILLIXR v4.0 is the first open-source end-to-end XR system that supports offloading of both head tracking and rendering (with 6DoF reprojection in the runtime), either separately or together. This release also introduces several other major upgrades, including hand tracking, plugin support for ORB_SLAM3 integration, and a new network backend that enables the offloading of multiple components under the coordination of the runtime.