Renshuai Tao | 中文简介

I am an Associate Professor in Institute of Information Science (Group Leader: Yao Zhao, IEEE Fellow), School of Computer and Information Technology, Beijing Jiaotong University. I received my B.S. and Ph.D. degree from Beihang University in 2017 and 2022, under supervision of Prof. Wei Li (academician of the Chinese Academy of Sciences) and Prof. Xianglong Liu. In my Ph.D. graduation ceremony, I was selected as the student representative to speak. After graduation, I worked as a Senior Research Engineer in Noah's Ark Lab of Huawei, working with Dr. Yunhe Wang. In Huawei's career, I was invited to participate in the "Spark Award" award ceremony and had a conversation with Mr. Zhengfei Ren. I serve as a reviewer for the top conferences and journals such as NeurIPS, CVPR, ICCV, ECCV, IJCAI, AAAI, MM, T-IFS, PR, etc.

Email: rstao [AT]


[Prospective students] Our group (Beijing Jiaotong University) has positions for PhD students, Master students, and visiting students. If you are interested, please feel free to send me an email with your CV and publications (if any).

Github / Google Scholar /


My research interests are broadly in

  • AI Safety and Security. This area mainly focuses on Digital Image Content Forensics, DeepFake Detection, Forensic Watermarking, etc. This field will be the new direction of my research and an important focus at BJTU.
  • Open-world Learning. This area mainly focuses on learning in some real-world scenerios, containing Open-set Learning, Domain Adaptation, Few-shot Learning, etc. The goal is to bridge the gap between learning on machines and humans.
  • Visual-language Models. Large-scale visual-language pre-training models have shown impressive transferability in many downstream tasks. This area brings a lot of fantastic works, e.g. CLIP and ALIGN, and there are still interesting works that need to be explored.
  • Interesting Evaluation Benchmarks. I like establish evaluation benchmarks for some very important areas that have not been studied recently. I hope these benchmarks can contribute to the community and encourage more researchers to continue to work in this field.

[2024.04] One paper is accepted by IJCAI 2024.

[2023.11] I join Beijing Jiaotong University as an Associate Professor.

[2023.02] One paper is accepted by USENIX Security Symposium 2023.

[2022.12] I am awarded the "Future Star" of HUAWEI company.

[2022.06] One first-authored paper is accepted by ACM Multimedia 2022.

[2022.06] I join Noah's Ark Lab, Huawei Company.

[2022.06] I am selected as the Student Representative to speak at the Graduation Ceremony.

[2022.06] I receive my Ph.D. degree.

[2022.03] Two papers including one first-authored are accepted by CVPR 2022.

[2021.10] I obtain the National Scholarship.

[2021.07] One first-authored paper for object detection is accepted by ICCV 2021.

Selected Papers | All Publications

Few-shot X-ray Prohibited Item Detection: A Benchmark and Weak-feature Enhancement Network.
Renshuai Tao, Tianbo Wang, Ziyang Wu, Cong Liu, Aishan Liu, Xianglong Liu.
ACM Multimedia, CCF-A, 2022
PDF / Code

We propose the first X-ray FSOD dataset on the typical industrial X-ray security inspection scenario consisting of 12,333 images and 41,704 instances from 20 categories, which could benchmark and promote FSOD studies on such more challenging scenarios.


Exploring Endogenous Shift for Cross-domain Detection: A Large-scale Benchmark and Perturbation Suppression Network.
Renshuai Tao, Hainan Li, Tianbo Wang, Yanlu Wei, Yifu Ding, Bowei Jin, Hongping Zhi, Xianglong Liu, Aishan Liu.
IEEE/CVF International Conference on Computer Vision and Pattern Recognition(CVPR), CCF-A, 2022
PDF / Code

We first put forward a novel and important type of domain shift in cross-domain detection, the endogenous shift, which may cause severe performance drop but has been rarely studied.


Towards Real-world X-ray Security Inspection: A High-Quality Benchmark And Lateral Inhibition Module For Prohibited Items Detection. [PDF]
Renshuai Tao, Yanlu Wei, Xiangjian Jiang, Hainan Li, Haotong Qin, Jiakai Wang, Yuqing Ma, Libo Zhang, Xianglong Liu.
IEEE International Conference on Computer Vision (ICCV), CCF-A, 2021
arXiv / Code

We present a High-quality X-ray (HiXray) security inspection image dataset and the Lateral Inhibition Module (LIM).


Over-sampling De-occlusion Attention Network for Prohibited Items Detection in Noisy X-ray Images [PDF]
Renshuai Tao, Yanlu Wei, Hainan Li, Aishan Liu, Yifu Ding, Haotong Qin, Xianglong Liu.
ArXiv, 2021
arXiv/ Code

To better improve occluded X-ray object detection, we further propose an over-sampling de-occlusion attention network (DOAM-O), which consists of a novel de-occlusion attention module and a new over-sampling training strategy.


Occluded Prohibited Items Detection: An X-ray Security Inspection Benchmark and De-occlusion Attention Module. [PDF]
Yanlu Wei*, Renshuai Tao*, Zhangjie Wu, Yuqing Ma, Libo Zhang, Xianglong Liu. (* indicates equal contribution)
ACM Multimedia (ACM MM), CCF-A,Oral, 2020
arXiv / Code

We contribute the first high-quality object detection dataset for security inspection, named Occluded Prohibited Items X-ray (OPIXray) image benchmark.


[2022.06]    Student Representative at the Graduation Ceremony (ONLY 1).

[2022.06]    Outstanding Graduates of Beijing Province.

[2022.01]    Beihang University Exploring Scholarship.

[2021.09]    National Scholarship (Top 2%).

[2021.06]    Beihang University Excellent Academic Paper Award.

[2021.01]    Outstanding Postgraduate of Beihang University (Top 10%).

[2020.11]    Merit Student of Beihang University.

[2020.11]    the First Prize of Excellent Postgraduate Scholarship of Beihang University (Top 20%).

[2019.10]    Beihang Ph.D. Scholarship for Outstanding Freshmen (Top 10%).


[Fall 2020]    Teaching Assistant in Machine Learning (Beihang University).

[Fall 2019]    Teaching Assistant in Machine Learning (Beihang University).