Hongsuk Benjamin Choi

image

Researcher

I am a Machine Learning / Computer Vision Researcher at Samsung AI center, New York. I received a Master’s degree at Seoul National University (SNU) computer vision lab after converting my PhD program. I was advised by Prof. Kyoung Mu Lee, Editor in Chief (EIC), IEEE Trans. on PAMI (Pattern Analysis and Machine Intelligence). Please refer to my CV for more information. Email: redstonepo@gmail.com

Professional: Google Scholar / Linkedin / Github
Casual but…: Youtube / Blog


Introduction

I am interested in a broad range of machine learning and computer vision topics. I have in-depth experience in 3D reconstruction and neural image generation.


Publications (Selected)

FineControlNet: Fine-level Text Control for Image Generation with Spatially Aligned Text Control Injection
Hongsuk Choi (*), Isaac Kasahara (*), Selim Engin, Moritz Alexander Graule, Nikhil Chavan-Dafle, Volkan Isler (* equal contribution)
preprint
[HOMEPAGE] [ARXIV] [CODE]




HandNeRF: Learning to Reconstruct Hand-Object Interaction Scene from a Single RGB Image
Hongsuk Choi, Nikhil Chavan-Dafle, Jiacheng Yuan, Volkan Isler, Hyunsoo Park
In ICRA 2024
[HOMEPAGE] [ARXIV] [VIDEO] [CODE]




Rethinking Self-Supervised Visual Representation Learning in Pre-training for 3D Human Pose and Shape Estimation
Hongsuk Choi (*), Hyeongjin Nam (*), Taeryung Lee, Gyeongsik Moon, Kyoung Mu Lee (* equal contribution)
In ICLR 2023
[ARXIV] [VIDEO]


Three Recipes for Better 3D Pseudo-GTs of 3D Human Mesh Estimation in the Wild
Gyeongsik Moon, Hongsuk Choi, Sanghyuk Chun, Jiyoung Lee, Sangdoo Yun
In CVPRW 2023
[ARXIV] [PDF] [HOMEPAGE]


MonoNHR: Monocular Neural Human Renderer
Hongsuk Choi (*), Gyeongsik Moon (*), Matthieu Armando, Vincent Leroy, Kyoung Mu Lee, Gregory Rogez (* equal contribution)
In 3DV 2022
[ARXIV] [VIDEO]


HandOccNet: Occlusion-Robust 3D Hand Mesh Estimation Network
JoonKyu Park (*), Yeonguk Oh (*), Gyeongsik Moon (*), Hongsuk Choi, Kyoung Mu Lee (* equal contribution)
In CVPR 2022
[ARXIV] [CODE]


Learning to Estimate Robust 3D Human Mesh from In-the-Wild Crowded Scenes
Hongsuk Choi, Gyeongsik Moon, JoonKyu Park, Kyoung Mu Lee
In CVPR 2022
[ARXIV] [CODE]


Accurate 3D Hand Pose Estimation for Whole-Body 3D Human Mesh Estimation
Gyeongsik Moon, Hongsuk Choi, Kyoung Mu Lee
In CVPRW 2022
[ARXIV] [VIDEO] [CODE]


NeuralAnnot: Neural Annotator for 3D Human Mesh Training Sets
Gyeongsik Moon, Hongsuk Choi, Kyoung Mu Lee
In CVPRW 2022
[ARXIV] [HOMEPAGE]


Beyond Static Features for Temporally Consistent 3D Human Pose and Shape from a Video
Hongsuk Choi, Gyeongsik Moon, Ju Yong Chang, Kyoung Mu Lee
In CVPR 2021
[ARXIV] [PDF] [CODE] [VIDEO]


Pose2Mesh: Graph Convolutional Network for 3D Human Pose and Mesh Recovery from a 2D Human Pose
Hongsuk Choi (*), Gyeongsik Moon (*) , Kyoung Mu Lee (* equal contribution)
In ECCV 2020
[ARXIV] [PDF] [CODE] [VIDEO]


Honors

SNU Distinguished Master Dissertation Award (2022)

1st place and 2nd place at the 3D human pose estimation in the wild (3DPW) challenge of a without association track in joint orientation and position metrics, respectively
(workshop in conjunction with ECCV 2020)

Qualcomm IT Tour 2019 presentation competition 1st place (2019)


Service


Experience

ML/CV Researcher, Samsung Research America, New York, NY, USA
(Jun. 2022 - present)

Visiting Researcher, Naver CLOVA AI Lab, Seoul, Korea
(Mar. 2022 - May. 2022)

Research Intern, Naver Labs Europe (NLE), Grenoble, France
(April. 2021 - Oct. 2021)

M.S., ECE, Seoul National University (SNU), Seoul, Korea
(Mar. 2020 - Feb. 2022)

Repulic of Korea Airforce (ROKA), Gyeryong, Korea
(Mar. 2015 - Mar. 2017)

B.S., Business, CSE, Seoul National University (SNU), Seoul, Korea
(Mar. 2013 - Feb. 2020)