Di Liu @ Rutgers

Brief Bio. I am a Ph.D. candidate in the Department of Computer Science at Rutgers University, advised by Board of Governors Professor Dimitris N. Metaxas. Prior to this, I was a Research Assistant in the Department of Electrical and Computer Engineering at The Johns Hopkins University.

My current research interests lie primarily in (1) 3D Computer Vision and Computer Graphics (e.g., Diffusion Models, Gaussian Splatting); (2) Large-scale multi-modality pretraining and foundation models for visual understanding (e.g., vision-language models, multimodal models, world models).

617 Bowser Rd
Piscataway, NJ 08854
CBIM, Rutgers

Email: di.liu [at] rutgers [dot] edu

Background

Education

Sept. 2021 - Present

Rutgers, The State University of New Jersey - New Brunswick

Ph.D. in Computer Science

Supervised by Prof. Dimitris N. Metaxas

Jan. 2020 - Aug. 2021

The Johns Hopkins University

Research Assistant. in Electrical and Computer Engineering

Supervised by Prof. Jerry L. Prince

Experience

Jan 2025 - Present

Snap Research NYC, Snap. New York City, NY, USA

Research Scientist Intern. Host: Jian Wang

Instant Generation of Animatable Dual-stylized Avatars. [under review]

May 2024 - Nov 2024

Codec Avatars Team, Reality Labs Research, Meta. Pittsburgh, PA, USA

Research Scientist Intern. Host: Chen Cao

Universal Layered Codec Avatar Reconstruction from Videos. [CVPR'25]

Jan. 2024 - May. 2024

Cognition, Cloud & AI, Microsoft. Redmond, WA

Research Scientist Intern. Host: Dongdong Chen

Large-scale Multi-modality Pretraining and Vision-language Models for Visual Understanding.

Jun. 2023 - Dec. 2023

Media Analytics Team, NEC Labs America, San Jose, CA, USA

Research Intern. Host: Bingbing Zhuang and Manmohan Chandraker

Instantaneous Perception of Moving Objects in 3D. [CVPR'24]

Sept. 2021 - Present

Rutgers University, New Brunswick, NJ, USA

Graduate Assistant. Supervised by: Prof. Dimitris N. Metaxas

Text-to-Character Blendshape Generation with Video Diffusion Models. [under review]
Layout-Agnostic Scene Text Image Synthesis with Diffusion Models. [CVPR'24]
3D Shape Reconstruction and Part Discovery from a Single Image. [ICCV'23] [NeurIPS'23]
Fundamental Research of Diffusion Models in Generative Modeling. [CVPR'25] , [ICLR'25]
Multi-modality Foundation Models in Medical Imaging. [CVPR'25] [CVPR'24]

Jan. 2020 - Aug. 2021

The Image Analysis and Communications Lab, The Johns Hopkins University, Baltimore, USA

Research Assistant. Host: Prof. Jerry L. Prince

Label Super Resolution for 3D Magnetic Resonance Images. [SPIE'21]

Selected Publications

(Please check Google Scholar for the full list of my publications.)

Snapmoji: Instant Generation of Animatable Dual-stylized Avatars
Di Liu^*, Eric Ming Chen^*, Sizhuo Ma, Michael Vasilkovsky, Bing Zhou, Qiang Gao, Wenzhou Wang, Jiahao Luo, Dimitris N. Metaxas, Vincent Sitzmann, Jian Wang
Under Review

[Paper] [Project Page] [Bibtex]

@misc{chen2025snapmojiinstantgenerationanimatable,
  title={Snapmoji: Instant Generation of Animatable Dual-Stylized Avatars}, 
  author={Eric M. Chen and Di Liu and Sizhuo Ma and Michael Vasilkovsky and Bing Zhou and Qiang Gao and Wenzhou Wang and Jiahao Luo and Dimitris N. Metaxas and Vincent Sitzmann and Jian Wang},
  year={2025},
  eprint={2503.11978},
  archivePrefix={arXiv},
  primaryClass={cs.GR},
  url={https://arxiv.org/abs/2503.11978}, 
}

LUCAS: Layered Universal Codec Avatars
Di Liu, Teng Deng, Giljoo Nam, Yu Rong, Stanislav Pidhorskyi, Junxuan Li, Jason Saragih, Dimitris N. Metaxas, Chen Cao
CVPR 2025

[Paper] [Project Page] [Bibtex]

    @inproceedings{liu2025lucas,
    title={LUCAS: Layered Universal Codec Avatars},
    author={Liu, Di and Deng, Teng and Nam, Giljoo and Rong, Yu and Pidhorskyi, Stanislav and Li, Junxuan and Saragih, Jason and Metaxas, Dimitris N. and Cao, Chen},
    booktitle={Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)},
    month={June},
    year={2025}
    }

T2Bs: Text-to-Character Blendshapes via Video Generation
Jiahao Luo, Chaoyang Wang, Michael Vasilkovsky, Vladislav Shakhrai, Di Liu, Peiye Zhuang, Sergey Tulyakov, Peter Wonka, Hsin-Ying Lee, James Davis, Jian Wang
Under Review

[Paper] [Project Page] [Bibtex]

Instantaneous Perception of Moving Objects in 3D
Di Liu, Bingbing Zhuang, Dimitris N. Metaxas, Manmohan Chandraker
CVPR 2024

[Paper] [Bibtex]

    @inproceedings{liu2024instantaneous,
      title={Instantaneous Perception of Moving Objects in 3D},
      author={Liu, Di and Zhuang, Bingbing and Metaxas, Dimitris N and Chandraker, Manmohan},
      booktitle={Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition},
      pages={19573--19583},
      year={2024}
    }

Layout-Agnostic Scene Text Image Synthesis with Diffusion Models
Qilong Zhangli, Jindong Jiang, Di Liu, Licheng Yu, Xiaoliang Dai, Ankit Ramchandani, Guan Pang, Dimitris N Metaxas, Praveen Krishnan
CVPR 2024

[Paper] [Bibtex]

@inproceedings{zhangli2024layout,
  title={Layout-Agnostic Scene Text Image Synthesis with Diffusion Models},
  author={Zhangli, Qilong and Jiang, Jindong and Liu, Di and Yu, Licheng and Dai, Xiaoliang and Ramchandani, Ankit and Pang, Guan and Metaxas, Dimitris N and Krishnan, Praveen},
  booktitle={Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition},
  pages={7496--7506},
  year={2024}
}

LEPARD: Learning Explicit Part Discovery for 3D Articulated Shape Reconstruction
Di Liu, Qilong Zhangli, Yunhe Gao, and Dimitris N Metaxas
NeurIPS 2023

[Paper] [Bibtex]

    @inproceedings{liu2023lepard,
    title={LEPARD: Learning Explicit Part Discovery for 3D Articulated Shape Reconstruction},
    author={Liu, Di and Zhangli, Qilong and Gao, Yunhe and Metaxas, Dimitris N},
    booktitle={Thirty-seventh Conference on Neural Information Processing Systems},
    year={2023}
    }

DeFormer: Integrating Transformers with Deformable Models for 3D Shape Abstraction from a Single Image
Di Liu, Xiang Yu, Meng Ye, Qilong Zhangli, Zhuowei Li, Zhixing Zhang, Dimitris N. Metaxas
ICCV 2023

[Paper] [Bibtex]

    @inproceedings{liu2023deformer,
    title={DeFormer: Integrating Transformers with Deformable Models for 3D Shape Abstraction from a Single Image},
    author={Liu, Di and Yu, Xiang and Ye, Meng and Zhangli, Qilong and Li, Zhuowei and Zhang, Zhixing and Metaxas, Dimitris N},
    booktitle={Proceedings of the IEEE/CVF International Conference on Computer Vision},
    pages={14236--14246},
    year={2023}
    }