I am a PhD candidate at National University of Singapore (NUS). My advisor is Prof. Mohan S Kankanhalli. I was an undergraduate of University of Science and Technology of China (USTC), where I graduated as an honored student of the Talent Program in Computer and Information Science and Technology.

I am interested in computer vision. In particular, I am interested in building the bridge between natural language, human knowledge, and the visual world. The goal of my research is to understand the complexities in the visual world utilizing compositional knowledge representations and model designs. My Google Scholar page is here. My CV is here.

My name is pronounced as zee-way-hsu, where Xu (hsu, 徐) is my family name and Ziwei (zee-way, 子威) is my given name.

Publications

Ziwei Xu, Yogesh S Rawat, Yongkang Wong, Mohan S Kankanhalli, and Mubarak Shah, Don’t Pour Cereal into Coffee: Differentiable Temporal Logic for Temporal Action Segmentation, NeurIPS, 2022. (25% acceptance rate) [Preprint] [Code] [Video]

Ziwei Xu, Xudong Shen, Yongkang Wong, and Mohan S Kankanhalli, Unsupervised Motion Representation Learning with Capsule Autoencoders, NeurIPS, 2021. (26% acceptance rate) [Preprint] [Code] [Slides] [Video]

Ziwei Xu, Guangzhi Wang, Yongkang Wong, and Mohan S Kankanhalli, Relation-aware Compositional Zero-shot Learning for Attribute-Object Pair Recognition, IEEE Transactions on Multimedia, 2021. [Preprint] [Code]

Junnan Li*, Ziwei Xu*, Yongkang Wong, Qi Zhao, and Mohan S Kankanhalli, GradMix: Multi-source Transfer across Domains and Tasks, WACV, 2020. [Preprint]

Yaqi Xie*, Ziwei Xu*, Mohan S Kankanhalli, Kuldeep Singh Meel, and Harold Soh, Embedding Symbolic Knowledge into Deep Networks, NeurIPS, 2019. (21% acceptance rate) [Preprint]

Ziwei Xu, Haitian Zheng, Minjian Pang, Xiongfei Su, Guyue Zhou, and Lu Fang, Utilizing High-Level Feature For Indoor Shopping Mall Localization, GlobalSIP, 2017.

Xiaoyun Yuan, Ziwei Xu, Haoqian Wang, Yebin Liu, and Lu Fang, Cascaded Image Deblurring With Combined Image Prior, GlobalSIP, 2017.

Haitian Zheng, Mengqi Ji, Ziwei Xu, Haoqian Wang, Yebin Liu, and Lu Fang, Learning Cross-scale Correspondence and Patch-based Synthesis for Reference-based Super-Resolution, BMVC, 2017.

* Equal contributions.

Community Services

  • PC Member: AAAI (since 2021), IJCAI (since 2021)