About me

I am a first-year PhD student from Center for Research in Computer Vision, University of Central Florida (UCF), supervised by Dr. Chen Chen.

My current research focuses on generative model (mainly on diffusion model now), such as, text to image generation, image restoration. Before joining UCF, I have spent three years working with Dr. Ziming Zhang at Worcester Polytechnic Institute, and obtained the Master degree of Electrical and Computer Engineering.

📝 Publications

Hyperbolic Chamfer Distance for Point Cloud Completion. ICCV 2023

Fangzhou Lin, Yun Yue, Songlin Hou, Xuechu Yu, Yajun Xu, Kazunori D Yamada, Ziming Zhang

• Apply Chamfer distance in Hyperbolic space to deal with outlier sensibility for Point Cloud Completion.
• Experimental results show this new method can get the state-of-the-art performance on the benchmark datasets, i.e. PCN, ShapeNet-55, and ShapeNet-34.

Leveraging Superfluous Information in Contrastive Representation Learning. arXiv 2024

Xuechu Yu, Fangzhou Lin, Yun Yue, Ziming Zhang

• Conjecture that the learned representations by contrastive learning not only maintain task-relevant information but also carry task-irrelevant information and design a new robust contrastive loss to eliminate this superfluous information.
• The new loss can often outperform the traditional contrastive learning approaches on image classification, object detection and instance segmentation tasks.

🎖 Honors and Awards

• UCF Graduate Artificial Intelligence Initiative Fellowship. 2024.
• China National Scholarship. 2015.
• China National Inspirational Scholarship. 2014.

📖 Educations

• 2024.09 - Now, Ph.D., Computer Science, University of Central Florida.
• 2021.09 - 2024.05, Master, Electrical and Computer Engineering, Worcester Polytechnic Institute.
• 2017.09 - 2020.06, Master, Electrical Engineering , Huazhong University of Science and Technology.
• 2013.09 - 2017.06, Bachelar, Electrical Engineering , Southwest Jiaotong University.