Automated semantic segmentation and object detection are of great importance in geospatial data analysis. However, supervised machine learning systems such as convolutional neural networks require large corpora of annotated training data. Especially in the geospatial domain, such datasets are quite scarce. Within this paper, we aim to alleviate this issue by introducing a new annotated 3D dataset that is unique in three ways: i) The dataset consists of both an Unmanned Aerial Vehicle (UAV) laser scanning point cloud and a 3D textured mesh. ii) The point cloud features a mean point density of about 800 pts/m2 and the oblique imagery used for 3D mesh texturing realizes a ground sampling distance of about 2–3 cm. This enables the identification of fine-grained structures and represents the state of the art in UAV-based mapping. iii) Both data modalities will be published for a total of three epochs allowing applications such as change detection. The dataset depicts the village of Hessigheim (Germany), henceforth referred to as H3D - either represented as 3D point cloud H3D(PC) or 3D mesh H3D(Mesh). It is designed to promote research in the field of 3D data analysis on one hand and to evaluate and rank existing and emerging approaches for semantic segmentation of both data modalities on the other hand. Ultimately, we hope that H3D will become a widely used benchmark dataset in company with the well-established ISPRS Vaihingen 3D Semantic Labeling Challenge benchmark (V3D). The dataset can be downloaded from https://ifpwww.ifp.uni-stuttgart.de/benchmark/hessigheim/default.aspx.
%0 Journal Article
%1 K_lle_2021
%A Kölle, Michael
%A Laupheimer, Dominik
%A Schmohl, Stefan
%A Haala, Norbert
%A Rottensteiner, Franz
%A Wegner, Jan Dirk
%A Ledoux, Hugo
%D 2021
%I Elsevier BV
%J ISPRS Open Journal of Photogrammetry and Remote Sensing
%K journal review
%P 100001
%R 10.1016/j.ophoto.2021.100001
%T The Hessigheim 3D (H3D) benchmark on semantic segmentation of high-resolution 3D point clouds and textured meshes from UAV LiDAR and Multi-View-Stereo
%U https://doi.org/10.1016%2Fj.ophoto.2021.100001
%V 1
%X Automated semantic segmentation and object detection are of great importance in geospatial data analysis. However, supervised machine learning systems such as convolutional neural networks require large corpora of annotated training data. Especially in the geospatial domain, such datasets are quite scarce. Within this paper, we aim to alleviate this issue by introducing a new annotated 3D dataset that is unique in three ways: i) The dataset consists of both an Unmanned Aerial Vehicle (UAV) laser scanning point cloud and a 3D textured mesh. ii) The point cloud features a mean point density of about 800 pts/m2 and the oblique imagery used for 3D mesh texturing realizes a ground sampling distance of about 2–3 cm. This enables the identification of fine-grained structures and represents the state of the art in UAV-based mapping. iii) Both data modalities will be published for a total of three epochs allowing applications such as change detection. The dataset depicts the village of Hessigheim (Germany), henceforth referred to as H3D - either represented as 3D point cloud H3D(PC) or 3D mesh H3D(Mesh). It is designed to promote research in the field of 3D data analysis on one hand and to evaluate and rank existing and emerging approaches for semantic segmentation of both data modalities on the other hand. Ultimately, we hope that H3D will become a widely used benchmark dataset in company with the well-established ISPRS Vaihingen 3D Semantic Labeling Challenge benchmark (V3D). The dataset can be downloaded from https://ifpwww.ifp.uni-stuttgart.de/benchmark/hessigheim/default.aspx.
@article{K_lle_2021,
abstract = {Automated semantic segmentation and object detection are of great importance in geospatial data analysis. However, supervised machine learning systems such as convolutional neural networks require large corpora of annotated training data. Especially in the geospatial domain, such datasets are quite scarce. Within this paper, we aim to alleviate this issue by introducing a new annotated 3D dataset that is unique in three ways: i) The dataset consists of both an Unmanned Aerial Vehicle (UAV) laser scanning point cloud and a 3D textured mesh. ii) The point cloud features a mean point density of about 800 pts/m2 and the oblique imagery used for 3D mesh texturing realizes a ground sampling distance of about 2–3 cm. This enables the identification of fine-grained structures and represents the state of the art in UAV-based mapping. iii) Both data modalities will be published for a total of three epochs allowing applications such as change detection. The dataset depicts the village of Hessigheim (Germany), henceforth referred to as H3D - either represented as 3D point cloud H3D(PC) or 3D mesh H3D(Mesh). It is designed to promote research in the field of 3D data analysis on one hand and to evaluate and rank existing and emerging approaches for semantic segmentation of both data modalities on the other hand. Ultimately, we hope that H3D will become a widely used benchmark dataset in company with the well-established ISPRS Vaihingen 3D Semantic Labeling Challenge benchmark (V3D). The dataset can be downloaded from https://ifpwww.ifp.uni-stuttgart.de/benchmark/hessigheim/default.aspx.},
added-at = {2021-09-22T11:11:52.000+0200},
author = {Kölle, Michael and Laupheimer, Dominik and Schmohl, Stefan and Haala, Norbert and Rottensteiner, Franz and Wegner, Jan Dirk and Ledoux, Hugo},
biburl = {https://puma.ub.uni-stuttgart.de/bibtex/2d6252b0b8e7f59258b0716ee508b4912/markusenglich},
doi = {10.1016/j.ophoto.2021.100001},
interhash = {8448e3a21a0d80bca52a92f9640fc2de},
intrahash = {d6252b0b8e7f59258b0716ee508b4912},
journal = {{ISPRS} Open Journal of Photogrammetry and Remote Sensing},
keywords = {journal review},
month = {10},
pages = 100001,
publisher = {Elsevier {BV}},
timestamp = {2021-12-02T15:04:43.000+0100},
title = {The Hessigheim 3D (H3D) benchmark on semantic segmentation of high-resolution 3D point clouds and textured meshes from {UAV} {LiDAR} and Multi-View-Stereo},
url = {https://doi.org/10.1016%2Fj.ophoto.2021.100001},
volume = 1,
year = 2021
}
