Xiao Fang

I am a Master's student in Robotics at Carnegie Mellon University. I am fortunate to be advised by Prof. Fernando De La Torre and to work with Prof. Shayok Chakraborty, and Dr. Stanislav Panev.

Previously, I received my Bachelor's degree in Computer Science and Mathematics at The Hong Kong University of Science and Technology. I was fortunate to be advised by Prof. Hao Chen and to work with Prof. Kwang-Ting Cheng, Prof. Dong Zhang, and Dr. Yi Lin.

Email  /  Scholar  /  LinkedIn  /  Twitter  /  Github

Research

I am broadly interested in computer vision. Currently, I am exploring how to leverage generative models to synthesize useful data for vision tasks. During my undergraduate study, I worked on developing efficient yet effective algorithms for medical image analysis tasks.

Leverage Stable Diffusion for weakly supervised cross-domain object detection and introduce two new datasets for vehicle detection in aerial imagery.

Propose Multi-Layer Permute Perceptron that integrates global interdependence in an efficient manner for volumetric medical image segmentation.

Propose StitchViT to efficiently capture long-range dependencies and explicitly use boundary information to guide medical image segmentation.

Translate abstract medical terminologies to visual symptoms that can be interpreted by Vision-Language Models.

Propose 2.5D UNet featuring spatial and squeeze attention for fat metaplasia segmentation on sacroiliac joint MRI.

Introduce Sobel Operator as an explicit boundary extractor to guide the model learning process for medical image segmentation.

Honors and Awards

Thanks Dr. Jon Barron for sharing the web template.