Jae Yong Lee
Ph.D in CS @ UIUC
I am a PhD candidate at the University of Illinois Urbana-Champaign (UIUC) specializing in computer vision. Additionally, I currently work as a computer vision engineer at Reconstruct.
My passion lies in the intersection of machine learning and computer vision, particularly in improving geometric computer vision problems with learned features. Recently, I have developed several innovative solutions, including a Real-time Neural Rendering system that eliminates the need for baking, an efficient neural scene representation, and a learning-based multi-view stereo system. I have also implemented segments of these algorithms in CUDA, where I have honed my expertise in estimating first-order and second-order gradients for complex function optimization.
In addition to my technical expertise, I am well-versed in turning ideas into scalable productions using cloud-based systems. My experience spans from server-side architecture, including load balancing and Dockerization, to delivery with Content Delivery Networks (CDN), all the way to client-side rendering with WebGL and GLSL.
Feel Free to reach out to me for any questions in my works!
I am mostly interested in solving geometric computer vision problems by using machine learning. Below are selected publications.
Meta Reality Lab