Homepage

Xin Li

Associate Research Fellow

Visual Intelligence Research Center

Pengcheng Laboratory

Email: xinlihitsz [at] gmail dot com

ABOUT ME

I am an associate research fellow at the Artificial Intelligence Center, Pengcheng Laboratory. Previously, I was a postdoctoral researcher at Pengcheng Laboratory working with Prof. Huchuan Lu. I obtained the Ph.D. degree at Harbin Institute of Technology, Shenzhen in 2020, under the advisement of Prof. Zhenyu He. From the fall of 2017 to the fall of 2018, I was supported by the China Scholarship Council and worked with Prof. Ming-Hsuan Yang as a joint-training Ph.D. student at the University of California, Merced. I am interested in computer vision and machine learning problems. Currently, I focus on developing Unified and Efficient Perception Models for Video Objects. (Google scholar)

NEWS

07/2025, we get 1st place in the Visual Object Tracking and Segmentation Challenge (VOTS25), ICCV 2025.
07/2025, our paper "Prior-Free Augmentation for Cloth-Changing Person Re-Identification " is accepted to ACM MM 2025.
01/2025, two papers are accepted to ICLR2025.
08/2024, we get 1st place in the Visual Object Tracking and Segmentation Challenge (VOTS24), ECCV 2024.
08/2024, we get 1st place in the Large-Scale Video Object Segmentation Challenge (LSVOS), VOS track, ECCV 2024.
06/2024, our paper "Spatial-Temporal Multi-level Association for Video Object Segmentation" is accepted to ECCV 2024.
06/2024, we get 1st place in the Pixel-level Video Understanding in the Wild Challenge, MOSE Track, CVPR 2024. Project page.
03/2024, our paper "Context-Guided Black-Box Attack for Visual Tracking" is accepted to TMM.
02/2024, our paper "RTracker: Recoverable Tracking via PN Tree Structured Memory" is accepted to CVPR 2024.
08/2023, we get 2nd place in the video object segmentation track in the ICCV Youtube VOS Challenge (2023).
07/2023, our paper "CiteTracker: Correlating Image and Text for Visual Tracking" is accepted to ICCV 2023.
12/2022, one self-supervised tracking paper is accepted to TNNLS.
09/2022, start my journey at PCL as an assistant research fellow.
03/2022, two papers are accepted to CVPR 2022.
01/2022, our paper "Learning Dual-Level Deep Representation for Thermal Infrared Tracking" is accepted to TMM.
08/2021, we release a large-scale neuromorphic tracking dataset VisEvent (paper).
07/2021, one visual tracking paper is accepted to ICCV 2021.
06/2021, two papers are accepted to TCSVT.
09/2020, our paper "Dual-regression model for visual tracking" is accepted to Neural Networks.
09/2020, start my postdoc at Artificial Intelligence Center, Peng Cheng Laboratory. Good luck!
07/2020, our paper "LSOTB-TIR: A Large-Scale High-Diversity Thermal Infrared Object Tracking Benchmark" is accepted to ACM International Conference on Multimedia 2020.
07/2020, our paper "Learning Deep Multi-Level Similarity for Thermal Infrared Object Tracking " is accepted to TMM.
01/2020, successfully defense my Ph.D. thesis.
11/2019, our paper "Multi-Task Driven Feature Models for Thermal Infrared Tracking'' is accepted to AAAI 2020.
03/2019, our paper "Target-Aware Deep Tracking'' is accepted to CVPR 2019.
09/2018, finish my visit at UCM and come back to China.
09/2017, start my visit at UCM.

SELECTED PUBLICATIONS

Spatial-Temporal Multi-level Association for Video Object Segmentation

Deshui Miao, Xin Li, Zhenyu He, Huchuan Lu, Ming-Hsuan Yang, ECCV, 2024.

[paper] [code]

RTracker: Recoverable Tracking via PN Tree Structured Memory

Yuqing Huang, Xin Li, Zikun Zhou, Yaowei Wang, Zhenyu He, Ming-Hsuan Yang, CVPR, 2024.

[paper] [code]

CiteTracker: Correlating Image and Text for Visual Tracking

Xin Li*, Yuqing Huang*, Zhenyu He, Yaowei Wang, Huchuan Lu, Ming-Hsuan Yang, ICCV, 2023.

[paper] [code]

Self-Supervised Tracking via Target-Aware Data Synthesis

Xin Li*, Wenjie Pei*, Yaowei Wang, Zhenyu He, Huchuan Lu, Ming-Hsuan Yang, TNNLS, 2023.

[paper] [code and raw results]

Multi-Objects Tracking Meets Moving UAV

Shuai Liu*, Xin Li*, Huchuan Lu, You He, CVPR, 2022.

[paper] [code]

Unsupervised Learning of Accurate Siamese Tracking

Qiuhong Shen, Lei Qiao, Jinyang Guo, Peixia Li, Xin Li, Bo Li, Weitao Feng, Weihao Gan, Wei Wu, Wanli Ouyang, CVPR, 2022.

[paper] [code]

Saliency-Associated Object Tracking

Zikun Zhou, Wenjie Pei, Xin Li, Hongpeng Wang, Feng Zheng, Zhenyu He, ICCV, 2021.

[paper] [code]

Dual-regression Model for Visual Tracking

Xin Li, Qiao Liu, Nana, Fan, Zikun Zhou, Zhenuyu He, Xiao-yuan Jing, Neural Networks, 2020.

[paper] [code]

LSOTB-TIR: a Large-scale High-diversity Thermal Infrared Object Tracking Benchmark

Qiao Liu, Xin Li, Zhenuyu He, Chenglong Li, Jun Li, Zikun Zhou, Di Yuan, Jing Li, Kai Yang, Nana Fan, Feng Zheng, in Twenty-Eighth ACM Multimedia Conference, (MM), 2020. .

[paper] [code and dataset]

Multi-Task Driven Feature Models for Thermal Infrared Tracking

Qiao Liu*, Xin Li*, Zhenuyu He, Nana Fan, Di Yuan, Wei Liu, Yongsheng Liang, in Thirty-Fourth AAAI Conference on Artifical Intelligence, (AAAI), 2020.

[paper] [code]