Reconstructing Compact Building Models from Point Clouds Using Deep Implicit Fields

Abstract

Three-dimensional (3D) building models play an increasingly pivotal role in many real-world applications, while obtaining a compact representation of buildings remains an open problem. In this paper, we present a novel framework for reconstructing compact, watertight, polygonal building models from point clouds. Our framework comprises three components: (a) a cell complex is generated via adaptive space partitioning that provides a polyhedral embedding as the candidate set; (b) an implicit field is learned by a deep neural network that facilitates building occupancy estimation; (c) a Markov random field is formulated to extract the outer surface of a building via combinatorial optimization. We evaluate and compare our method with state-of-the-art methods in shape reconstruction, surface approximation, and geometry simplification. Experiments on both synthetic and real-world point clouds have demonstrated that, with our neural-guided strategy, high-quality building models can be obtained with significant advantages in fidelity, compactness, and computational efficiency. Our method shows robustness to noise and insufficient measurements, and it can directly generalize from synthetic scans to real-world measurements.

Deep Learning for 3d Building Reconstruction: A Review

“…The holistic primitive fitting method (Zhang et al, 2021) was also used along with PointNet++ (Qi et al, 2017) for 3D building reconstruction from point clouds. Another three-step 3D building reconstruction approach using deep implicit fields and point clouds was proposed by Chen et al (2021). The DL methods also can be combined with GIS for reconstructing 3D city models from high-resolution satellite imagery (Pepe et al, 2021).…”

Implicity: City Modeling From Satellite Images With Deep Implicit Occupancy Fields

“…The network can be evaluated at any 3D coordinate and, therefore, conceptually, allows for infinite resolution-in practice, its effective resolution is bounded by the representation power of the finite number of neurons, as well as by the resolution of the training data. So far, implicit representations have been explored to model the 3D geometry of local shapes (Genova et al, 2019, Genova et al, 2020, single objects (Park et al, 2019, Atzmon andLipman, 2020), indoor scenes (Jiang et al, 2020, Peng et al, 2020, Sitzmann et al, 2020, Chabra et al, 2020, and single buildings (Chen et al, 2021). In this work, we go one step further and investigate their potential to accurately reconstruct 3D urban scenes, on the order of several km 2 , from satellite data.…”