Qirui Wu

qiruiw at sfu dot ca

self.jpeg

I am a Ph.D student in 3DLG Lab at Simon Fraser University, supervised by Prof. Angel X. Chang. I also closely collaborate with Prof. Manolis Savva and Prof. Daniel Ritchie. My research goal aims at designing systems and developing algorithms for understanding and modeling cluttered lifelike 3D scenes that enables interactions among embodied agents and humans. More specifically, I’m interested in

  • Multimodal 3D foundation models unifying understanding and generation
  • Compositional and dynamic 3D content generation ranging from object parts to large scenes
  • General 3D vision in robotics and human-centric AI

Google Scholar / Twitter / Github

Research

2025

  1. diorama.png
    Diorama: Unleashing Zero-shot Single-view 3D Scene Modeling
    Qirui Wu, Denys Iliash, Daniel Ritchie, Manolis Savva, and Angel X. Chang
    In Proc. of the International Conference on Computer Vision (ICCV), 2025
    arXiv Code Website

2024

  1. r3ds.png
    R3DS: Reality-linked 3D Scenes for Panoramic Scene Understanding
    Qirui Wu, Sonia Raychaudhuri, Daniel Ritchie, Manolis Savva, and Angel X. Chang
    In Proc. of the European Conference on Computer Vision (ECCV), 2024
    arXiv Code Website
  2. generalizing_shape_retrieval.png
    Generalizing Single-View 3D Shape Retrieval to Occlusions and Unseen Objects
    Qirui Wu, Daniel Ritchie, Manolis Savva, and Angel X. Chang
    In Proc. of the International Conference on 3D Vision (3DV), 2024
    arXiv Code Website

2022

  1. d3net.png
    D^3Net: A Unified Speaker-Listener Architecture for 3D Dense Captioning and Visual Grounding
    Dave Zhenyu Chen, Qirui Wu, Matthias Niessner, and Angel X. Chang
    In Proc. of the European Conference on Computer Vision (ECCV), 2022
    arXiv PDF Code Website

2021

  1. plan2scene.png
    Plan2Scene: Converting Floorplans to 3D Scenes
    Madhawa Vidanapathirana, Qirui Wu, Yasutaka Furukawa, Angel X. Chang, and Manolis Savva
    In Proc. of the Conference on Computer Vision and Pattern Recognition (CVPR), 2021
    arXiv Code Website