Final year Ph.D. student @KAIST bio-imaging signal processing & learning lab (BISPL). Prior research intern at NVIDIA Research, Google Research Perception (LUMA), and Los Alamos National Laboratory (LANL) applied math and plasma physics group (T-5). Diffusion models and inverse problems enthusiast. Hyungjin Chung has pioneered and advanced some of the most widely acknowledged works on diffusion model-based inverse problem solvers. My research interests include, but are not restricted to advancing and widening the applicability of diffusion models in inverse imaging, applications to solve real-world problems (e.g. medical imaging).
Download my CV.
PhD in Bio & Brain Engineering, Current
KAIST
MS in Bio & Brain Engineering, 2021
KAIST
BS in Biomedical Engineering, 2019
Korea University
By generalizing DIP, we can design an adaptation algorithm that corrects the PF-ODE trajectory of posterior sampling with diffusion models, such that one can reconstruct from OOD measurements.
Prompt tuning of text embedding leads to better reconstruction quality when solving inverse problems with latent diffusion models.
DDS enables fast sampling from the posterior without the need for heavy gradient computation in DIS.
We show that seemingly different direct diffusion bridges are equivalent, and that we can push the pareto frontier of the perception-distortion tradeoff with data consistency gradient guidance.
TPDM improves 3D voxel generative modeling with 2D diffusion models. We show that 3D generative prior can be accurately represented as the product of two independent 2D diffusion priors that scale to both unconditional sampling and solving inverse problems.
We propose a method that can solve 3D inverse problems in the medical imaging domain using only the pre-trained 2D diffusion model augmented with the conventional model-based prior.
Diffusion posterior sampling enables solving arbitrary noisy (e.g. Gaussian, Poisson) inverse problems that are both linear or non-linear.
Manifold constraint dramatically improves the performance of unsupervised inverse problem solving using diffusion models.
Come-close to diffuse-fast when solving inverse problems with diffusion models. We establish state-of-the-art results with only 20 diffusion steps across various tasks including SR, inpainting, and CS-MRI