Sejong Yang

Student, Developer, and Researcher


2019.09 - present
I am a Ph.D. Candidate at Yonsei University, supervised by Prof. Seon Joo Kim. I have explored super-resolution with lightweight deep neural network(network pruning). Currently, I am very much interested in talking face motion modeling with generative models(VAE, GAN, Normalizing Flows, Diffusion Models), 3D model(3DMM, Neural Renderer, NeRF), and multi-modal representation(visual-audio, visual-text).
2018.07 - 2019.08
As an undergraduate intern in the lab of Seon Joo Kim(My Ph.D. advisor), I try to find video highlights in baseball games automatically with video summarization methods.
I engage in the start-up Blammedia as a CTO in 2019. We got the preliminary start-up package investment worth 30M won from the Korea Institute of Startup and Entrepreneurship Development. I learn the market value of technology and product by meeting with venture capitalists.
2018.01 - 2018.06
I spent a wonderful spring semester as an exchange student at Uppsala University in 2018. Learning computer science in different culture was a great experience.
2016.11 - 2017.11
At Ewoosoft, I participate as a developer of EzMobile, Ez3D-i. I learn how to use Xamarin.Forms, Qt5, and Flask for developing the enterprise-level product. I also become familiar with collaboration/devops tools like Jira, Atlassian, Crucible and Fisheye, Github, and Jenkins with the waterfall development model.
2016.08 - 2016.11
I have software assistant experience at Kakao in 2016 developing a component for openstack. In the experience, I learn the basics of the network, virtualization, and cloud computing from the experience.
2014.03 - 2016.08
BEng at Yonsei University. This is the place where I know what is "coding" for the first time, and I managed to get a degree in computer science from it in the end. As an undergraduate intern in the lab of Seon Joo Kim(My Ph.D. advisor), I conducted for collecting and purifying data for color constancy task with the hyespectral camera.
Publications
Dense Interspecies Face Embedding
NeurIPS 2022
Sejong Yang, Subin Jeon, Seonghyeon Nam, and Seon Joo Kim.
[paper] [code] [project page]
SRFlow-DA: Super-Resolution Using Normalizing Flow with Deep Convolutional Block
CVPRW 2021
Younghyun Jo, Sejong Yang, and Seon Joo Kim.
[paper] [code]
Investigating loss functions for extreme super-resolution
CVPRW 2020
Younghyun Jo, Sejong Yang, and Seon Joo Kim.
[paper] [code]
Miscellanies