Jeongsoo Park
jp2678 at cornell dot edu

I am a third-year CS PhD student at Cornell Tech, advised by Prof. Andrew Owens. I have a general interest in computer vision, image processing, and visual forensics.

I was a CSE PhD candidate at the University of Michigan before I moved with my advisor to Cornell Tech on Fall 2025.

I graduated with a master's degree in ECE from the University of Michigan, during which I was advised by Prof. Justin Johnson.

I received a BS degree in Electrical Engineering from Sungkyunkwan University, where I was advised by Prof. Jong Hwan Ko.

LinkedIn  /  Google Scholar  /  Twitter  /  Github

I have been working on an intersection of computer vision, image processing, and image forensics.

In my recent work Community Forensics, I study the generalization of fake image detectors by training them using images sampled from thousands of generators. In RGB no more, I train Vision Transformers directly from the encoded features of a JPEG image to accelerate the entire pipeline. In Auto-Tiler, I design an efficient compression scheme for over-the-network model inference.

We train fake image detectors using images from thousands of various generators to study generalization in detecting generated images.

We train ViTs directly from JPEG encoded features and accelerate train and eval by up to 39.2% and 17.9%.

We use autoencoders and show up to 67% higher accuracy and 81% reduced latency during network-constrained inference versus the image or video codec-based compression.

This website uses the html template by Jon Barron.