Building Roof Vectorization With Ppgnet

Abstract

Abstract. A challenge in data-based 3D building reconstruction is to find the exact edges of roof facet polygons. Although these edges are visible in orthoimages, convolution-based edge detectors also find many other edges due to shadows and textures. In this feasibility study, we apply machine learning to solve this problem. Recently, neural networks have been introduced that not only detect edges in images, but also assemble the edges into a graph. When applied to roof reconstruction, the vertices of the dual graph represent the roof facets. In this study, we apply the Point-Pair Graph Network (PPGNet) to orthoimages of buildings and evaluate the quality of the detected edge graphs. The initial results are promising, and adjusting the training parameters further improved the results. However, in some cases, additional work, such as post-processing, is required to reliably find all vertices.

Roof3d: A Real and Synthetic Data Collection for Individual Building Roof Plane and Building Sections Detection

“…Name Area Representation Ground Truth Modalities RoofWorld (Nauata and Furukawa, 2020) single building planes RGB RoofVec (Hensel et al, 2021) 2.00 km 2 single building planes RGB ISPRS Potsdam 3.42 km 2 tiles footprints RGB+IR+DSM SemCity (inst) (Roscher et al, 2020) 3.02 km 2 tiles sections RGB UBC (Huang et al, 2022b) 66.12 km 2 patch sections RGB Roof3D (ours)…”

“…The City3D (Huang et al, 2022a) dataset contains 20,000 building instances in LoD-2 along with airborne lidar point-clouds. In the two works of (Nauata and Furukawa, 2020) and (Hensel et al, 2021) datasets for roof geometry extraction along with ortho-imagery are provided, but DSM data is not included.…”

Evaluating Convnet and Transformer Based Self-Supervised Algorithms for Building Roof Form Classification

“…For our experiments, we conducted linear evaluation using building roof vectorization dataset published by Hensel et al (2021). However, the dataset lacked roof type information.…”