Senior Research Scientist — 3D Shape Modeling & Processing
Minseo Park
Current focus: Real-time drivable Gaussian-Splat scene reconstruction at 30 Hz for closed-loop simulation.
Background
PhD Stanford CS 2019, advised by Leonidas J. Guibas in the Geometric Computing group. Thesis on implicit neural representations for partial point clouds. Research Scientist at NVIDIA Toronto AI Lab (2019–2024) under Sanja Fidler, working on driving-scene reconstruction via Neuralangelo and 3D Gaussian Splatting. Joined Blueberry to lead the 3D scene-modeling pillar.
Education
- PhD · Computer Science — Stanford University (2019), advisor: Leonidas J. Guibas
- MS · Computer Science — KAIST (2014)
Selected publications
- Park et al., 'Latent Occupancy Fields for Partial Point Cloud Completion', CVPR 2020
- Park et al., 'Neural Signed Distance Fields for Driving Scenes', ECCV 2022
- Park et al., 'Drivable 3D Gaussian Splats from Calibrated Camera Rigs', SIGGRAPH 2024
Reading list
- Mescheder et al., 'Occupancy Networks', CVPR 2019
- Mildenhall et al., 'NeRF: Representing Scenes as Neural Radiance Fields for View Synthesis', ECCV 2020
- Kerbl et al., '3D Gaussian Splatting for Real-Time Radiance Field Rendering', SIGGRAPH 2023
- Yang et al., 'EmerNeRF: Emergent Spatial-Temporal Scene Decomposition via Self-Supervision', ICLR 2024
- Tonderski et al., 'NeuRAD: Neural Rendering for Autonomous Driving', CVPR 2024