Ziyu
0%

Edge XR Latency

Posted on In

Conference Papers

  1. [VRST]Predictability-Aware Motion Prediction for Edge XR via High-Order Error-State Kalman Filtering
    ACM Symposium on Virtual Reality Software and Technology (VRST), ACM, 2025.
    Authors: Ziyu Zhong, …
    Status: Published
    Summary: Latency-aware motion prediction for distributed / cloud XR rendering, compensating motion-to-photon latency using SLAM-based and ML-based pose prediction.
    DOI: 10.1145/3756884.3765973

Ongoing Research Projects

  1. [AIXR]
    IEEE Conference on Artificial Intelligence for Extended Reality (AIXR), IEEE, Year.
    Authors: Ziyu Zhong, …
    Status: Accepted
    Summary: Robust multimodal head-pose tracking for XR HMDs, fusing OpenXR, IMU/visual signals, and mocap under unknown extrinsics and clock drift.
    Link: (add IEEE Xplore / DOI link here)

  2. Low-Latency AR with LLM Backend (ChatGPT-style)

  • Designing an AR system that integrates head/hand pose and visual context with LLM responses.
  • Profiling communication and inference bottlenecks, and exploring predictive scheduling to reduce perceived latency.
  • Investigating how multimodal signals (motion + camera + language) can be combined to drive responsive AR assistance.