Xiaoshuai Zhang

UC San Diego email: zхs АТ uсsd DОТ еdu

I'm a first-year Ph.D. Student working with Prof. Hao Su at the Dept. of Computer Science and Engineering, UC San Diego. My goal is to empower robots to see and perceive like us human beings do, with the most advanced computer vision technologies. Specifically, I’m interested in 3D Vision, Low-Level Vision and problems related to Multi-View Geometry. I received my B.S. in Machine Intelligence from Peking University, advised by Professor Jiaying Liu. I also worked with Haoqiang Fan at Face++ while I was an undergraduate.

For leisure, I love travelling and experiencing different cultures. I am also an amateur photographer (see my flickr). To English speakers: Shall-shy Chaung is a good approximation for my name (张孝帅, IPA: /ʈ͡ʂɑŋ ɕiau ʂuai/). You can also call me by my nickname Jet.



1. Disentangled Image Matting

Shaofan Cai*, Xiaoshuai Zhang*, Haoqiang Fan, Haibin Huang, Jiangyu Liu, Jiaming Liu, Jiaying Liu, Jue Wang, and Jian Sun

Accepted by the International Conference on Computer Vision (ICCV), 2019. [PDF] [arXiv] [Code (soon)]

2. Dynamically Unfolding Recurrent Restorer: A Moving Endpoint Control Method for Image Restoration

Xiaoshuai Zhang*, Yiping Lu*, Jiaying Liu and Bin Dong.

Accepted by the International Conference on Learning Representations (ICLR), 2019. [OpenReview] [Project Page] [Code]

3. How You Act Tells a Lot: Privacy-Leaking Attack on Deep Reinforcement Learning

Xiaoshuai Zhang*, Weiyao Wang*, Xinlei Pan*, Bo Li, Jinfeng Yi, and Dawn Song.

Accepted by the AAMAS, 2019. [arXiv]

4. Bridging the Gap Between Computational Photography and Visual Recognition

Submitted to some journal. [arXiv]

5. A Comprehensive Benchmark for Single Image Compression Artifacts Reduction

Submitted to some journal. [arXiv]


1. DMCNN: Dual-Domain Multi-scale Convolutional Neural Network for Compression Artifacts Removal

Xiaoshuai Zhang, Wenhan Yang, Yueyu Hu and Jiaying Liu.

Accepted by the IEEE International Conference on Image Processing (ICIP), 2018 as an oral presentation. [Paper] [arXiv] [Project Page] [code]