Nanjing UniversityI'm now a Ph.D student in MCG Group, Computer Science, Nanjing University, under the supervision of Asst. Jie Liu, Assoc. Prof. Jie Tang, Prof. Limin Wang and Prof. Gangshan Wu. Before that, I also spent wonderful years as an undergraduate in the Department of Computer Science and Technology, Nanjing University, and received my Bachelor of Science in Jun 2025.
My research interests include Computer Vision and Deep Learning. I have a long-standing focus on Multiple Object Tracking, and my current work explores Multimodal Learning and Generative Models.
Education
-
Nanjing UniversityDepartment of Electrical Engineering and Computer Science
Ph.D. StudentSep. 2025 - present -
Nanjing UniversityB.S. in Computer ScienceSep. 2021 - Jul. 2025
Experience
-
Taobao and Tmall Group, AlibabaJava Software EngineerJun. 2024 - Sep. 2024
Honors & Awards
-
People's Scholarship2024
-
People's Scholarship2023
-
Agilent Scholarship2022
News
Selected Publications (view all )
GLAD: Generative Language-Assisted Visual Tracking for Low-Semantic Templates
Xingyu Luo, Yidong Cai, Jie Liu, Jie Tang, Gangshan Wu, Limin Wang
International Journal of Computer Vision (IJCV-CCF A) 2026
Proposed GLAD, a generative language-assisted visual tracking framework that introduces a diffusion-based generative multimodal fusion paradigm to effectively integrate textual descriptions with low-semantic template images, enhancing cross-modal alignment and semantic representation, and achieving state-of-the-art performance on multiple benchmarks.
GLAD: Generative Language-Assisted Visual Tracking for Low-Semantic Templates
Xingyu Luo, Yidong Cai, Jie Liu, Jie Tang, Gangshan Wu, Limin Wang
International Journal of Computer Vision (IJCV-CCF A) 2026
Proposed GLAD, a generative language-assisted visual tracking framework that introduces a diffusion-based generative multimodal fusion paradigm to effectively integrate textual descriptions with low-semantic template images, enhancing cross-modal alignment and semantic representation, and achieving state-of-the-art performance on multiple benchmarks.