Gwanghyun (Bradley) Kim

About Me

I am a Research Scientist at NVIDIA GEAR, working on robot foundation models and world models. My research focuses on unifying perception, prediction, and action for real-world embodied agents, building on my background in multimodal generative modeling of the physical world.

I received my Ph.D. in Electrical and Computer Engineering from Seoul National University, advised by Prof. Se Young Chun, and my M.S. in Bio and Brain Engineering from KAIST, advised by Prof. Jong Chul Ye. During my Ph.D., I interned at NVIDIA Research and Google Research. My research has been supported by the Qualcomm Innovation Fellowship.

News

[Apr. 2026] I joined NVIDIA GEAR as a Research Scientist.
[Feb. 2026] HUG-3D is accepted to CVPR 2026 as a Highlight.
[Sep. 2025] We won the 2nd place award in REG 2025.
[Jun. 2025] GeoMan (NVIDIA) and PersonaCraft are accepted to ICCV 2025.
[Oct. 2024] I won the Google Conference Scholarship.
[Sep. 2024] AVDiT & MoNL (Google) is accepted to NeurIPS 2024.
[Jul. 2024] BeyondScene is accepted to ECCV 2024.
[Jan. 2024] I joined NVIDIA as a Research Intern.
[Nov. 2023] I won the Qualcomm Innovation Fellowship South Korea 2023.
[Sep. 2023] I joined Google as a Student Researcher.

[Aug. 2023] I am selected as a recipient of Yulchon AI Star Scholarships.
[Jul. 2023] PODIA-3D is accepted to ICCV 2023.
[Apr. 2023] Ditto-NeRF is available on arXiv.
[Feb. 2023] DATID-3D is accepted to CVPR 2023.
[Aug. 2022] DiffusionCLIP is invited to oral session at KCCV 2022.
[Jun. 2022] DISTL is accepted to Nature Communications.
[Mar. 2022] DiffusionCLIP is accepted to CVPR 2022.
[Dec. 2021] I won the Best Master Student for the Year Award at BISPL, KAIST.
[Sep. 2021] FeSTA is accepted to NeurIPS 2021.
[Dec. 2020] We won the 2nd place award in NVIDIA AI Healthcare Hackathon.

Selected Research

World Action Models are Zero-Shot Policies

S. Ye*, Y. Ge, K. Zheng, S. Gao, S. Yu, ..., Gwanghyun Kim, ..., Y. Chebotar, S. Reed, J. Kautz, Y. Zhu*, J. Fan*, J. Jang* (*leads)

arXiv 2026, NVIDIA

PDF Project page Code
CVPR

Human Interaction-Aware 3D Reconstruction from a Single Image

Gwanghyun Kim*, J. J. Kim*, S. Y. Jeon*, J. Park, S. Y. Chun (*co-first authors)

CVPR 2026 (Highlight)

PDF Project page
ICCV

GeoMan: Temporally Consistent Human Geometry Estimation using Image-to-Video Diffusion

Gwanghyun Kim, X. Li, Y. Yuan, K. Nagano, T. Li, J. Kautz, S. Y. Chun, U. Iqbal

ICCV 2025, NVIDIA

PDF Project page
ICCV

PersonaCraft: Personalized and Controllable Full-Body Multi-Human Scene Generation Using Occlusion-Aware 3D-Conditioned Diffusion

Gwanghyun Kim*, S. Y. Jeon*, S. Lee, S. Y. Chun (*co-first)

ICCV 2025

PDF Project page Code
NeurIPS

A Versatile Diffusion Transformer with Mixture of Noise Levels for Audiovisual Generation

Gwanghyun Kim*, A. Martinez, Y. Su, B. Jou, J. Lezama, A. Gupta, L. Yu, L. Jiang, A. Jansen, J. Walker, K. Somandepalli* (*co-first)

NeurIPS 2024, Google

PDF Project page
ECCV

BeyondScene: Higher-Resolution Human-Centric Scene Generation With Pretrained Diffusion

Gwanghyun Kim*, H. Kim*, H. Seo*, D U Kang*, S. Y. Chun (*co-first)

ECCV 2024

PDF Project page Code
ICCV

PODIA-3D: Domain Adaptation of 3D Generative Model Across Large Domain Gap Using Pose-Preserved Text-to-Image Diffusion

Gwanghyun Kim, J. H. Jang, S. Y. Chun

ICCV 2023

PDF Project page Code
CVPR

DATID-3D: Diversity-Preserved Domain Adaptation Using Text-to-Image Diffusion for 3D Generative Model

Gwanghyun Kim, S. Y. Chun

CVPR 2023

PDF Project page Code Demo
DITTO-NeRF: Diffusion-based Iterative Text To Omni-directional 3D Model

H. Seo*, H. Kim*, Gwanghyun Kim*, S. Y. Chun (*co-first)

arxiv 2023

PDF Project page Code
CVPR

DiffusionCLIP: Text-Guided Diffusion Models for Robust Image Manipulation

Gwanghyun Kim, T. Kwon, J. C. Ye

CVPR 2022

PDF Code Demo
Nat. Commun.

AI can evolve without labels: self-evolving vision transformer for chest X-ray diagnosis through knowledge distillation

S. Park, Gwanghyun Kim, Y. Oh, J. B. Seo, S. M. Lee, J. H. Kim, S. Moon, J. K. Lim, C. M. Park, J. C. Ye

Nature Communications, 2022

PDF
NeurIPS

Federated Split Vision Transformer for COVID-19 CXR Diagnosis using Task-Agnostic Training

S. Park*, Gwanghyun Kim*, J. Kim, B. Kim, J. C. Ye (*co-first)

NeurIPS 2021

PDF
Med. Image Anal.

Multi-task Vision Transformer using Low-level Chest X-ray Feature Corpus for COVID-19 Diagnosis and Severity Quantification

S. Park*, Gwanghyun Kim*, Y. Oh, J. B. Seo, S. M. Lee, J. H. Kim, S. Moon, J. K. Lim, J. C. Ye (*co-first)

Medical Image Analysis, 2021

PDF