Semantic 3D urban mesh benchmark



Summary

Recent advances in Structure from Motion (SfM) and Multi-View Stereo (MVS) enable generating textured meshes of large scale urban scenes. Many real-world applications (e.g., sunlight simulation) require detailed semantic information of these models. With the current state-of-the-art machine learning / deep learning techniques, it is possible to perform semantic segmentation on urban meshes with the feeding of large annotated data sets.

The project aims to build benchmark data sets for semantic 3D urban meshes and a mesh annotation platform.


Annotation software

Coming soon …

Data sets

Coming soon …

Funding

EuroSDR logo

This project has received funding from EuroSDR.


Team

Weixiao  Gao photo

Weixiao Gao
PhD Candidate

|

Hugo  Ledoux photo

Hugo Ledoux
Associate-prof.

| | |

Liangliang  Nan photo

Liangliang Nan
Assistant-prof

|

Jantien  Stoter photo

Jantien Stoter
Professor

| | |

Student Helper

Ziqian Ni, assists in software development, from 2019-07 to 2019-09.
Mels Smit, assists in mesh annotation, from 2020-07 to 2020-09.
Charalampos Chatzidiakos, assists in mesh annotation, from 2020-07 to 2020-09.