
Decoupled Reprojection Consistency for Diagnosing 3D Gaussian Splatting Failures
Eurographics 2026 Poster
I am currently a research intern at the Computational Light Laboratory, University College London, working with Prof. Kaan Akşit. My background includes industry experience in game production and AI-driven content pipelines. My research interests focus on perceptual graphics, inverse rendering, and neural 3D representations. My goal is to develop representations of light and visual content that account for human perception. I am currently seeking PhD opportunities in these areas.
I received my M.S. in Computer Science from Chung-Ang University, where I worked in the Automated Machine Learning Lab on a range of machine learning topics.
I was credited on Lies of P (2023), an AAA title developed by NEOWIZ.

Eurographics 2026 Poster

Neowiz, Jan 2024 - Present
An internal generative image service for controllable, title-specific artwork creation across multiple game projects.

Neowiz, Apr 2024 - Nov 2024
A production-oriented pipeline for converting dense reconstructed hair into editable guide-hair models.

Neowiz, Aug 2022 - Oct 2022
An automated facial animation pipeline that generated lip-sync and emotional expressions from voice and dialogue scripts.

Neowiz, Jun 2021 - Oct 2022
Conducted GAN-based domain transfer to convert monster voices into mechanical sounds; applied in Lies of P.

National Research Foundation of Korea, Sep 2018 - Oct 2019
Conducted 3D volumetric data classification using 3D convolutional neural networks, achieving high accuracy in multi-class prediction tasks.
I draw inspiration and motivation from visual things. Visual computing and painting are how I give form to my reflections.