Jacob joined the lab as a Ph.D. student in 2019, after a B.S. in Astrophysics and Statistics from the University of Toronto. He works on 3D point cloud understanding with deep learning, including semantic segmentation, upsampling and diffusion models.
Current projects
- YUPU: airborne LiDAR point-cloud upsampling benchmark (ECCV 2026 Workshop)
- Multimodal point-cloud segmentation for transmission corridor mapping
- EyeNet++ (IEEE TGRS 2025)
Publications (8)
2026
- Yoo, S. and Sohn, G. (2026). Appearance-aware Scaling Diffusion Model for 3D Point Cloud Upsampling. Int. Arch. Photogramm. Remote Sens. Spatial Inf. Sci., XLIX-B2-2026. doi:10.5194/isprs-archives-XLIX-B2-2026-315-2026
- Korny, Y., Yoo, S., and Sohn, G. (2026). Polarization-Aware Segmentation for Camouflaged Threat Detection from UAVs. Int. Arch. Photogramm. Remote Sens. Spatial Inf. Sci., XLIX-B2-2026. doi:10.5194/isprs-archives-XLIX-B2-2026-705-2026
- Korny, Y., Yoo, S., Panangian, D., Bittner, K., Wichmann, A., and Sohn, G. (2026). GeoPriorPC: Nadir-view to 3D Point Cloud Reconstruction for Buildings via Two-Stage Diffusion Priors. Proceedings of the IEEE/CVF CVPR Workshops.
2025
- Yoo, S., Jeong, Y., Sheikholeslami, M. M., and Sohn, G. (2025). EyeNet++: A Multiscale and Multidensity Approach for Outdoor 3-D Semantic Segmentation Inspired by the Human Visual Field. IEEE Transactions on Geoscience and Remote Sensing, 63, 1-19.
2023
- Yoo, S., Jeong, Y., Jameela, M., and Sohn, G. (2023). Human Vision Based 3D Point Cloud Semantic Segmentation of Large-Scale Outdoor Scenes. IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops, 6577-6586.
- Jameela, M., Sohn, G., and Yoo, S. (2023). Fusion-SUNet: Spatial Layout Consistency for 3D Semantic Segmentation. IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops, 6568-6576.
- Yoo, S., Ko, C., Sohn, G., and Lee, H. (2023). YUTO Semantic: A Large Scale Aerial LiDAR Dataset for Semantic Segmentation. Int. Arch. Photogramm. Remote Sens. Spatial Inf. Sci.
Preprints and Under Review
- Alizadeh Naeini, A., Sheikholeslami, M. M., Ghorbanalivakili, M., Yoo, S., and Sohn, G. A Unified Multi-Task Learning Framework for DTM Generation Using Surface Differencing. SSRN 5045901.