Jaesung Choe
Research Scientist at NVIDIA
Seoul, South Korea
Summary
👤
Senior🎓
Top SchoolJaesung Choe is a research scientist based in Seoul with eight years of experience advancing 3D computer vision and neural rendering, currently leading research on 3D multimodal large-language models at NVIDIA. He has a strong academic foundation from KAIST (PhD and MS) and practical industry experience from internships at Meta and NVIDIA focused on point-based NeRF, dynamic scenes, and video-based reconstruction. Jaesung mentors PhD interns, fosters academic collaborations, and has contributed to cutting-edge projects like 3D Gaussian Splatting that bridge research and real-world graphics pipelines. Known for combining deep theoretical grounding with hands-on systems work, he frequently publishes and shares work through his personal homepage and GitHub.
8 years of coding experience
Bachelor's degree, Bachelor's degree at Sogang UniversityDoctor of Philosophy - PhD, Doctor of Philosophy - PhD at Korea Advanced Institute of Science and Technology
English, cuda, pytorch
Github Skills (106)
python10
image-processing10
deep-learning10
instance-segmentation10
gpu10
mask-rcnn10
mask10
object-detection10
3d-reconstruction10
pointnet10
machine-learning10
keras10
segmentation10
rendering9
convolutional-neural-networks9
Programming languages (5)
Github contributions (5)
LifeBeyondExpectations/ECCV22-PointMixer
Jul 2022 - Dec 2022
[ECCV 2022] Official pytorch implementation of the paper, "PointMixer: MLP-Mixer for Point Cloud Understanding"
Contributions:120 commits, 15 PRs, 60 pushes in 5 months
mlpmlp-mixereccveccv2022point-cloud
LifeBeyondExpectations/PointMixer
Dec 2021 - Jul 2022
Contributions:16 commits, 15 pushes, 1 branch in 7 months
Find and Hire Top DevelopersWe’ve analyzed the programming source code of over 60 million software developers on GitHub and scored them by 50,000 skills. Sign-up on Prog,AI to search for software developers.