Jiayi Liu

I am a Ph.D. candidate at Simon Fraser University in GrUVi & 3DLG lab. I am honored to be co-supervised by Prof. Arash (Ali) Mahdavi-Amiri and Prof. Manolis Savva. Before this, I received my M.Sc. degree in computer science from University of Southern California.

A central question in my work is how to bridge visual observation, 3D geometry, functional structure, and physical interaction. Moving beyond static 3D shapes, I am especially interested in moving beyond static 3D shapes toward 3D representations that capture how objects are structured, how they move, and how they support interaction in everyday environments.

In the long term, I hope to contribute to interactive world modeling for robotics and embodied AI. I am excited by the possibility of building virtual environments where agents can learn to interact with the world through physically meaningful actions.

jiayi.jpg

Publications

2026

  1. artiverse.png
    Artiverse: A Diverse and Physically Grounded Dataset for Articulated Objects
    In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2026

2025

  1. survey.png
    Survey on Modeling of Human-made Articulated Objects
    Jiayi LiuManolis Savva, and Ali Mahdavi-Amiri
    In Eurographics STAR (State of The Art Report), Computer Graphics Forum, 2025
  2. singapo.png
    SINGAPO: Single Image Controlled Generation of Articulated Parts in Objects
    In Proceedings of the International Conference on Learning Representations (ICLR), 2025

2024

  1. cage.webp
    CAGE: Controllable Articulation GEneration
    In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2024

2023

  1. iccv23_paris.gif
    PARIS: Part-level Reconstruction and Motion Analysis for Articulated Objects
    Jiayi LiuAli Mahdavi-Amiri, and Manolis Savva
    In Proceedings of the IEEE International Conference on Computer Vision (ICCV), 2023

2022

  1. cvpr22_exemplar.png
    Exemplar-based Pattern Synthesis with Implicit Periodic Field Network
    In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2022

2021

  1. iccv21_tofu.png
    Topologically Consistent Multi-View Face Inference Using Volumetric Sampling
    In Proceedings of the IEEE International Conference on Computer Vision (ICCV), 2021