About

I am a PhD student in the Division of Industrial Data Science at the School of Data Science, Lingnan University, Hong Kong. My research is on the mathematical foundations of image segmentation, the task of dividing an image into meaningful regions.

Classical methods such as the Mumford–Shah functional define segmentation through explicit energies with a solid variational theory. Modern deep learning models such as Segment Anything learn from large annotated datasets and often perform better, but they lack a clear mathematical description and produce unstructured sets of masks.

My work aims to bridge the two approaches, develop variational formulations that cover both, and I use learned features within classical models to combine the strength of learning with the rigour and structure of explicit energies.

Publications

Publications will be listed here soon.