Weikai (Wei-Kai) Lin

Department of Computer Science @ University of Rochester, NY, USA

lwk2.jpg
LinkedIn / Google Scholar / GitHub

I am a Ph.D. student at University of Rochester, advised by Yuhao Zhu. Previously, I earned B.E. from Tsinghua University and M.Sc. from Peking University. Find more details in my CV (updated 11/02/2025).
đź’» My current research aims to bridge the gap between hardware and software through system-level co-optimization, including but not limited to:

  • 🚀 AR/VR Neural Rendering and Acceleration: Developing neural rendering algorithms for AR/VR systems and designing corresponding hardware accelerators. [SIGGRAPH Asia'25, ASPLOS'25a, ASPLOS'25b, ISCA'25, HotMobile'25, TACO'24]
  • đź“· Imaging System Design: Designing sensors and optics for imaging systems, along with co-optimized downstream vision models. [DAC'25, WACV'25]

👉 If you are interested in collaboration, Feel free to reach out via email: wlin33 at ur.rochester.edu.

News🔥

Sep 19, 2025 Our paper, PowerGS, accepted to Siggraph Asia 2025. See you in Hong Kong.
Apr 01, 2025 Our paper, Metasapiens, received ASPLOS’25 Best Paper Award.
Mar 22, 2025 Our paper, Lumina, accepted to ISCA 2025.
Feb 15, 2025 Our paper, SnapPix, accepted to DAC 2025.
Jan 31, 2025 Awarded the OpenAI Research Access Grant.
Jan 28, 2025 Our paper, StreamGrid, accepted to ASPLOS 2025.
Dec 16, 2024 Our paper on streaming 3DGS accepted to HotMobile 2025.
Oct 28, 2024 Our paper, PrivateEye, accepted to WACV 2025.
Oct 02, 2024 Our paper, Metasapiens, accepted to ASPLOS 2025.

Selected publicationsđź“ť

( * Indicates Equal contributions.)
  1. SIGGRAPH Asia
    PowerGS: Display–Rendering Power Co-Optimization for Neural Rendering in Power-Constrained XR Systems
    Weikai Lin, Sushant Kondguli, Carll Marshall, and 1 more author
    In SIGGRAPH Asia 2025 Conference Papers, 2025
    arXiv HTML Code
  2. ArXiv
    ControlHair: Physically-based Video Diffusion for Controllable Dynamic Hair Rendering
    Weikai Lin, Haoxiang Li, and Yuhao Zhu
    In arXiv:2509.21541, 2025
    arXiv HTML
  3. DAC
    SnapPix: Efficient-Coding–Inspired In-Sensor Compression for Edge Vision
    Weikai Lin*, Tianrui Ma*, Adith Boloor, and 4 more authors
    In 62nd ACM/IEEE Design Automation Conference, 2025
    arXiv Code
  4. ASPLOS
    MetaSapiens: Real-Time Neural Rendering with Efficiency-Aware Pruning and Accelerated Foveated Rendering
    🏆 ASPLOS Best Paper Award
    Weikai Lin*, Yu Feng*, and Yuhao Zhu
    In 30th ACM International Conference on Architectural Support for Programming Languages and Operating Systems, 2025
    🏆 ASPLOS’25 Best Paper Award HTML PDF Code
  5. ASPLOS
    StreamGrid: Streaming Point Cloud Analytics via Compulsory Splitting and Deterministic Termination
    Yu Feng*, Zheng Liu*, Weikai Lin*, and 6 more authors
    In 30th ACM International Conference on Architectural Support for Programming Languages and Operating Systems, 2025
    arXiv HTML
  6. ISCA
    Lumina: Real-Time Mobile Neural Rendering by Exploiting Computational Redundancy
    Yu Feng, Weikai Lin, Yuge Cheng, and 6 more authors
    In 52nd IEEE/ACM International Symposium on Computer Architecture, 2025
    arXiv HTML
  7. WACV
    Private-Eye: In-Sensor Privacy Preservation Through Optical Feature Separation
    Adith Boloor, Weikai Lin, Tianrui Ma, and 3 more authors
    In IEEE/CVF Winter Conference on Applications of Computer Vision, 2025
    PDF
  8. HotMoblie
    Advancing Immersive Content Delivery with Dynamic 3D Gaussian Splatting
    Nan Wu*, Weikai Lin*, Ruizhi Cheng*, and 4 more authors
    In 26th International Workshop on Mobile Computing Systems and Applications, 2025
    HTML
  9. TACO
    Potamoi: Accelerating Neural Rendering via a Unified Streaming Architecture
    Yu Feng*, Weikai Lin*, Zihao Liu, and 6 more authors
    In ACM Transactions on Architecture and Code Optimization, 2024
    PDF
  10. IROS
    OW3Det: Toward Open-World 3D Object Detection for Autonomous Driving
    Wenfei Hu, Weikai Lin, Hongyu Fang, and 2 more authors
    In IEEE/RSJ International Conference on Intelligent Robots and Systems, 2024
    PDF

Visits 🌍