Wentao Cheng

Associate Professor

Beijing Normal-Hong Kong Baptist University

Email: wentaocheng@bnbu.edu.cn

I am an Associate Professor at Beijing Normal-Hong Kong Baptist University (BNBU).

Prior to joining BNBU in Feb 2025, I worked as an Associate Professor at Nankai University (2022–2024) and as a Senior Algorithm Engineer at Alibaba Group AI Lab (2020–2022).

I received my Joint Ph.D. degree in Computer Science and Engineering from Nanyang Technological University (NTU), Singapore, and Technische Universität Darmstadt, Germany. I obtained my B.Eng. degree from Harbin Institute of Technology (HIT).

My research interests lie in 3D Computer Vision, with a particular focus on:

Visual Localization
3D Reconstruction
Efficient Feed-Forward 3D Foundation Models

I actively serve as a reviewer for top-tier conferences and journals, including CVPR, ICCV, ECCV, ICRA, TIP, and TRO. Additionally, I serve as an Executive Committee Member of the Technical Committee on CAD and Computer Graphics (CAD/CG), China Computer Federation (CCF).

Students who are interested in my research topics, please refer to this and contact me.

news

Sep 07, 2026	Congratulations to my two FYP groups on having their papers accepted to ICVISP 2026!
Jul 08, 2026	Our new arXiv preprint “ReCal3R: Reliability-Calibrated Learning Rates for Streaming 3D Reconstruction” is now available. Check out the paper, project page, and code.
May 01, 2026	One paper titled “QuadBox: Accelerating 3D Gaussian Splatting with Geometry-Aware Boxes” was accepted to ICIP 2026. This is the second first-author paper by our undergraduate student.
Mar 18, 2026	Two papers on 3D Foundation Models and Local Feature Learning have been accepted by ICME 2026! 🎉🎉 S-VGGT: Structure-Aware Subscene Decomposition for Scalable 3D Foundation Models. D2Feat: Dual Distillation for Semantic and Geometric Local Feature Learning. Stay tuned for the full papers!
Dec 16, 2025	I served as the Lead Organizer for the CCF CAD&CG Academic Seminar (BNBU Station). We invited four distinguished experts to discuss Visual Perception and Graphics.

selected publications

arXiv

ReCal3R: Reliability-Calibrated Learning Rates for Streaming 3D Reconstruction

Xinze Li, Yiyuan Wang, Pengxu Chen, Weifeng Su, Weisi Lin, and Wentao Cheng^*

arXiv preprint arXiv:2607.05356, 2026

arXiv Bib PAGE Code

@article{li2026recal3r,
  title = {ReCal3R: Reliability-Calibrated Learning Rates for Streaming 3D Reconstruction},
  author = {Li, Xinze and Wang, Yiyuan and Chen, Pengxu and Su, Weifeng and Lin, Weisi and Cheng, Wentao},
  journal = {arXiv preprint arXiv:2607.05356},
  year = {2026},
}

ICIP

QuadBox: Accelerating 3D Gaussian Splatting with Geometry-Aware Boxes

Xinze Li, Bohan Yang, Pengxu Chen, Yiyuan Wang, Hongcheng Luo, Weifeng Su, and Wentao Cheng^*

In IEEE International Conference on Image Processing (ICIP), 2026

Bib PDF Code

@inproceedings{li2026quadbox,
  title = {QuadBox: Accelerating 3D Gaussian Splatting with Geometry-Aware Boxes},
  author = {Li, Xinze and Yang, Bohan and Chen, Pengxu and Wang, Yiyuan and Luo, Hongcheng and Su, Weifeng and Cheng, Wentao},
  booktitle = {IEEE International Conference on Image Processing (ICIP)},
  year = {2026},
}

ICME

S-VGGT: Structure-Aware Subscene Decomposition for Scalable 3D Foundation Models

Xinze Li, Pengxu Chen, Yiyuan Wang, Weifeng Su, and Wentao Cheng^*

In IEEE International Conference on Multimedia and Expo (ICME), 2026

Bib PDF Code

@inproceedings{li2026svggt,
  title = {S-VGGT: Structure-Aware Subscene Decomposition for Scalable 3D Foundation Models},
  author = {Li, Xinze and Chen, Pengxu and Wang, Yiyuan and Su, Weifeng and Cheng, Wentao},
  booktitle = {IEEE International Conference on Multimedia and Expo (ICME)},
  year = {2026},
}

ICME

D2Feat: Dual Distillation for Semantic and Geometric Local Feature Learning

Yiyuan Wang, Xinze Li, Puzhen Wu, Junkai Zhang, Weifeng Su, and Wentao Cheng^*

In IEEE International Conference on Multimedia and Expo (ICME), 2026

Bib PDF

@inproceedings{wang2026d2feat,
  title = {D2Feat: Dual Distillation for Semantic and Geometric Local Feature Learning},
  author = {Wang, Yiyuan and Li, Xinze and Wu, Puzhen and Zhang, Junkai and Su, Weifeng and Cheng, Wentao},
  booktitle = {IEEE International Conference on Multimedia and Expo (ICME)},
  year = {2026},
}

ICVISP

Efficient Mamba Feature Matching via Block-Routed Sparse Coarse Matching

Boyu Liu, Jianxu Tang, Haoyu Xiong, and Wentao Cheng^*

In International Conference on Vision, Image and Signal Processing (ICVISP 2026), 2026

Bib

@inproceedings{liu2026efficient,
  title = {Efficient Mamba Feature Matching via Block-Routed Sparse Coarse Matching},
  author = {Liu, Boyu and Tang, Jianxu and Xiong, Haoyu and Cheng, Wentao},
  booktitle = {International Conference on Vision, Image and Signal Processing (ICVISP 2026)},
  year = {2026},
}

ICVISP

SAKA: Spatially Aware Keypoint Allocation for Lightweight Image Matching

Haoran Zhao, Leyan Chen, Haokai Wen, and Wentao Cheng^*

In International Conference on Vision, Image and Signal Processing (ICVISP 2026), 2026

Bib

@inproceedings{zhao2026saka,
  title = {SAKA: Spatially Aware Keypoint Allocation for Lightweight Image Matching},
  author = {Zhao, Haoran and Chen, Leyan and Wen, Haokai and Cheng, Wentao},
  booktitle = {International Conference on Vision, Image and Signal Processing (ICVISP 2026)},
  year = {2026},
}

ICCVW

Beyond Flat Text: Dual Self-inherited Guidance for Visual Text Generation

Minxing Luo, Zixun Xia, Liaojun Chen, Zhenhang Li, Weichao Zeng, Jianye Wang, Wentao Cheng, Yaxing Wang, Yu Zhou, and Jian Yang

In IEEE/CVF International Conference on Computer Vision (ICCV) HiGen Workshop, 2025

Bib PDF

@inproceedings{luo2025beyond,
  title = {Beyond Flat Text: Dual Self-inherited Guidance for Visual Text Generation},
  author = {Luo, Minxing and Xia, Zixun and Chen, Liaojun and Li, Zhenhang and Zeng, Weichao and Wang, Jianye and Cheng, Wentao and Wang, Yaxing and Zhou, Yu and Yang, Jian},
  booktitle = {IEEE/CVF International Conference on Computer Vision (ICCV) HiGen Workshop},
  year = {2025},
}

CVPR

ExtDM: Distribution Extrapolation Diffusion Model for Video Prediction

Zhicheng Zhang, Junyao Hu, Wentao Cheng^*, Danda Paudel, and Jufeng Yang

In IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2024

Bib PDF Code

GitHub stars for nku-zhichengzhang/ExtDM

@inproceedings{zhang2024extdm,
  title = {ExtDM: Distribution Extrapolation Diffusion Model for Video Prediction},
  author = {Zhang, Zhicheng and Hu, Junyao and Cheng, Wentao and Paudel, Danda and Yang, Jufeng},
  booktitle = {IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)},
  year = {2024},
}

SPL

MVP: One-Shot Object Pose Estimation by Matching with Visible Points

Wentao Cheng and Minxing Luo

IEEE Signal Processing Letters (SPL), 2024

Bib PDF Code

@article{cheng2024mvp,
  title = {MVP: One-Shot Object Pose Estimation by Matching with Visible Points},
  author = {Cheng, Wentao and Luo, Minxing},
  journal = {IEEE Signal Processing Letters (SPL)},
  year = {2024},
}

RA-L

Road Mapping and Localization Using Sparse Semantic Visual Features

Wentao Cheng, Sheng Yang, Maomin Zhou, Ziyuan Liu, Yiming Chen, and Mingyang Li

IEEE Robotics and Automation Letters (RA-L), 2021

Bib PDF

@article{cheng2021road,
  title = {Road Mapping and Localization Using Sparse Semantic Visual Features},
  author = {Cheng, Wentao and Yang, Sheng and Zhou, Maomin and Liu, Ziyuan and Chen, Yiming and Li, Mingyang},
  journal = {IEEE Robotics and Automation Letters (RA-L)},
  year = {2021},
}

ICCV

Cascaded Parallel Filtering for Memory-Efficient Image-Based Localization

Wentao Cheng, Weisi Lin, Kan Chen, and Xinfeng Zhang

In IEEE International Conference on Computer Vision (ICCV), 2019

Bib PDF Code

GitHub stars for wentaocheng-cv/cpf_localization

@inproceedings{cheng2019cascaded,
  title = {Cascaded Parallel Filtering for Memory-Efficient Image-Based Localization},
  author = {Cheng, Wentao and Lin, Weisi and Chen, Kan and Zhang, Xinfeng},
  booktitle = {IEEE International Conference on Computer Vision (ICCV)},
  year = {2019},
}

TIP

A Two-Stage Outlier Filtering Framework for City-Scale Localization Using 3D SfM Point Clouds

Wentao Cheng, Kan Chen, Weisi Lin, Michael Goesele, Xinfeng Zhang, and Yabin Zhang

IEEE Transactions on Image Processing (TIP), 2019

Bib PDF

@article{cheng2019two,
  title = {A Two-Stage Outlier Filtering Framework for City-Scale Localization Using 3D SfM Point Clouds},
  author = {Cheng, Wentao and Chen, Kan and Lin, Weisi and Goesele, Michael and Zhang, Xinfeng and Zhang, Yabin},
  journal = {IEEE Transactions on Image Processing (TIP)},
  year = {2019},
}

TIP

A Data-Driven Point Cloud Simplification Framework for City-Scale Image-Based Localization

Wentao Cheng, Weisi Lin, Xinfeng Zhang, Michael Goesele, and Ming-Ting Sun

IEEE Transactions on Image Processing (TIP), 2017

Bib

@article{cheng2016data,
  title = {A Data-Driven Point Cloud Simplification Framework for City-Scale Image-Based Localization},
  author = {Cheng, Wentao and Lin, Weisi and Zhang, Xinfeng and Goesele, Michael and Sun, Ming-Ting},
  journal = {IEEE Transactions on Image Processing (TIP)},
  year = {2017},
}