Publications

Do you want to hear about a highlight of my work, Scan2LoD3 method? -> listen to the podcast ;)

2024


ZAHA: Introducing the Level of Facade Generalization and the large-scale point cloud facade semantic segmentation benchmark dataset

Wysocki, O., Tan Y., Froech T., Xia, Y., Wysocki M., Hoegner, L., Cremers D., and Holst Ch

Accepted for WACV ‘25

[arxiv]


FacaDiffy: Inpainting unseen facade parts using diffusion models

Froech, T., Wysocki, O., Xia, Y., Xie, J., Schwab, B., Cremers D., Kolbe, TH.

Under review for ISPRS Annals

[own preprint]


Mind the domain gap: Measuring the domain gap between real-world and synthetic point clouds for automated driving development

Duc, N., Lai, YL., Madlindl, P., Zhu, X., Schwab, B., Wysocki, O., Hoegner, L., Kolbe, TH

Under review for PFG

[own preprint]


CDGS: Confidence-aware depth regularization for 3D Gaussian Splatting

Zhang, Q., Wysocki, O., Urban, S., Jutzi, B.

ISPRS Archives, LowCost 3D ‘24

[website] [pdf]


Analyzing the impact of semantic LoD3 building models on image-based vehicle localization

Bieringer, A., Wysocki, O., Tuttas, S., Hoegner, L., Holst, Ch.

ISPRS Annals, 3DGeoInfo ‘24

[website] [pdf] [arxiv]


Enriching thermal point clouds of buildings using semantic 3D building models

Zhu, J., Wysocki, O., Holst, Ch., Kolbe, TH. ,

ISPRS Annals, 3DGeoInfo ‘24

[website] [pdf] [arxiv]


Reconstructing facade details using MLS point clouds and Bag-of-Words approach

Froech, T., Wysocki, O., Hoegner, L., and Stilla, U.

Recent Advances in 3D Geoinformation Science, 3D GeoInfo ‘23

[website] [pdf] [arxiv]


Classifying point clouds at the facade-level using geometric features and deep learning networks

Tan, Y., Wysocki, O., Hoegner, L., and Stilla, U.

Recent Advances in 3D Geoinformation Science, 3D GeoInfo ‘23

[website] [pdf] [arxiv]


Transferring facade labels between point clouds with semantic octrees while considering change detection

Schwarz, S., Pilz, T., Wysocki, O., Hoegner, L., and Stilla, U.

Recent Advances in 3D Geoinformation Science, 3D GeoInfo ‘23

[website] [pdf] [arxiv]


MLS2LoD3: Refining low LoDs building models with MLS point clouds to reconstruct semantic LoD3 building models

Wysocki, O., Hoegner, L., and Stilla, U.

Recent Advances in 3D Geoinformation Science, 3D GeoInfo ‘23

[website] [pdf] [arxiv]

2023


Scan2LoD3: Reconstructing semantic 3D building models at LoD3 using ray casting and Bayesian networks

Wysocki, O., Xia, Y., Wysocki M., Grilli, E., Hoegner, L., Cremers D., and Stilla, U.

CVPRW ‘23

[website] [pdf] [arxiv]

2022


Combining visibility analysis and deep learning for refinement of semantic 3D building models by conflict classification

Wysocki, O., Grilli, E., Hoegner, L. and Stilla, U.

ISPRS Annals, 3D GeoInfo ‘22

[website] [pdf] [arxiv]


Refinement of semantic 3D building models by reconstructing underpasses from MLS point clouds

Wysocki, O., Hoegner, L. and Stilla, U.

International Journal of Applied Earth Observation and Geoinformation

[website] [pdf]


TUM-FAÇADE: Reviewing and enriching point cloud benchmarks for façade segmentation

Wysocki, O., Hoegner, L. and Stilla, U.

ISPR Archives, ArCH ‘22

[website] [pdf] [arxiv][Open3D-ML dataset]

2021


Unlocking point cloud potential: Fusing MLS point clouds with semantic 3D building models while considering uncertainty

Wysocki, O., Xu, Y. and Stilla, U.

ISPRS Annals, 3D GeoInfo ‘21

[website] [pdf]


Plastic surgery for 3D city models: A pipeline for automatic geometry refinement and semantic enrichment

Wysocki, O., Schwab, B., Hoegner, L., Kolbe, TH. and Stilla, U.

ISPRS Annals, ISPRS Congress ‘21

[website] [pdf]