Om Khangaonkar

I'm a fourth-year undergrad at UC Davis advised by Hamed Pirsiavash. My research studies computer vision and machine learning.

Specifically, I am interested in the intersection of representation learning, generative modeling, and scene understanding. Large generative models (i.e. Stable Diffusion, FLUX.1) have learned to model a large amount of our visual world. How can we utilize the rich representations learned by these models to build generalizable models of perception from limited supervision, similar to humans?

Email  /  Twitter  /  Google Scholar

We finetune generative models (i.e. Stable Diffusion, MAE) to segment object instances for a narrow set of object types. We find many interesting properties emerge including 1) zero-shot generalization to objects nothing like finetuning data 2) excellent performance at segmenting fine structures 3) very precise object edges.

Thanks to Jon Barron for this website template.