About me

I am a researcher with Shanghai AI Laboratory, working in a research group on Content Generation and Digitization. I received my Ph.D. (2018-2022) from Multimedia Laboratory (MMLab) at CUHK, advised by Prof.Dahua Lin. I obtained my Bachelor's Degree (2014-2018) at Xi'an Jiaotong University.
My current research focuses on 3D content generation, including 3D object and scene generation. I am looking for highly motivated interns in 3D generative models at Shanghai AI Laboratory. Drop me an email (lvzhaoyang@pjlab.org.cn) if you are interested.

Education

    The Chinese University of Hong Kong (CUHK)

    August 2018 - July 2022

    Ph.D. in Information Engineering

    Xi'an Jiaotong University (XJTU)

    August 2014 - July 2018

    B.S. in Physics (Experimental Class)

Publications

3D Content Generation

    GetMesh: A Controllable Model for High-quality Mesh
    Generation and Manipulation

    Zhaoyang Lyu, Ben Fei, Jinyi Wang, Xudong Xu, Ya Zhang, Weidong Yang, Bo Dai

    arXiv preprint

    [Paper] [Code] [Bibtex]

    SLIDE: Controllable Mesh Generation Through Sparse Latent
    Point Diffusion Models

    Zhaoyang Lyu, Jinyi Wang, Yuwei An, Ya Zhang, Dahua Lin, Bo Dai

    IEEE Conference on Computer Vision and Pattern Recognition (CVPR ) 2023

    [Paper] [Code] [Bibtex]

    A Conditional Point Diffusion-Refinement Paradigm for 3D
    Point Cloud Completion

    Zhaoyang Lyu, Zhifeng Kong, Xudong Xu, Liang Pan, Dahua Lin

    International Conference on Learning Representations (ICLR ) 2022

    [Paper] [Code] [Bibtex]

    MATLABER: Material-Aware Text-to-3D via LAtent BRDF auto-EncodeR

    Xudong Xu, Zhaoyang Lyu, Xingang Pan, Bo Dai

    arXiv preprint

    [Paper] [Code] [Bibtex]

Diffusion Model

    Accelerating Diffusion Models via Early Stop of the Diffusion Process

    Zhaoyang Lyu, Xudong Xu, Ceyuan Yang, Dahua Lin, Bo Dai

    arXiv preprint

    [Paper] [Code] [Bibtex]

    Generative Diffusion Prior for Unified Image Restoration and Enhancement

    Ben Fei, Zhaoyang Lyu (Equal Contribution), Liang Pan, Junzhe Zhang, Weidong Yang, Tianyue Luo,
    Bo Zhang, Bo Dai

    IEEE Conference on Computer Vision and Pattern Recognition (CVPR ) 2023

    [Paper] [Code] [Bibtex]

    DiffBIR: Towards Blind Image Restoration with Generative Diffusion Prior

    Xinqi Lin, Jingwen He, Ziyan Chen, Zhaoyang Lyu, Ben Fei, Bo Dai, Wanli Ouyang, Yu Qiao, Chao Dong

    arXiv preprint

    [Paper] [Code] [Bibtex]

    Point Cloud Pre-training with Diffusion Models

    Xiao Zheng, Xiaoshui Huang, Guofeng Mei, Yuenan Hou, Zhaoyang Lyu, Bo Dai,
    Wanli Ouyang, Yongshun Gong

    IEEE Conference on Computer Vision and Pattern Recognition (CVPR ) 2024

    [Paper] [Code (Coming Soon)] [Bibtex]

Neural Network Robustness

    POPQORN: Quantifying Robustness of Recurrent Neural Networks

    Ching-Yun Ko, Zhaoyang Lyu (Equal Contribution), Lily Weng, Luca Daniel, Ngai Wong, Dahua Lin

    International Conference on Machine Learning (ICML ) 2019

    [Paper] [Code] [Bibtex]

    Fastened CROWN: Tightened Neural Network Robustness Certificates

    Zhaoyang Lyu, Ching-Yun Ko, Zhifeng Kong, Ngai Wong, Dahua Lin, Luca Daniel

    AAAI Conference on Artificial Intelligence (AAAI ) 2020

    [Paper] [Code] [Bibtex]

    Towards Evaluating and Training Verifiably Robust Neural Networks

    Zhaoyang Lyu, Minghao Guo, Tong Wu, Guodong Xu, Kehuan Zhang, Dahua Lin

    IEEE Conference on Computer Vision and Pattern Recognition (CVPR ) 2021

    [Paper] [Code] [Bibtex]

    Guided Diffusion Model for Adversarial Purification

    Jinyi Wang, Zhaoyang Lyu (Equal Contribution), Dahua Lin, Bo Dai, Hongfei Fu

    arXiv preprint

    [Paper] [Code] [Bibtex]