Global, Dense Multiscale Reconstruction for a Billion Points

75 citations recorded by the source corpus.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Octree Generating Networks: Efficient Convolutional Architectures for High-resolution 3D Outputs (2017, background): “Octrees have a long history in classical 3D reconstruction and depth map fusion [21, 1, 11, 34, 6, 31].”
EC-Net: an Edge-aware Point set Consolidation Network (2018, methodology): “We also include a recent method for large-scale surface reconstruction [52].”
DeMoN: Depth and Motion Network for Learning Monocular Stereo (2016, methodology): “We used the Citywall and Achteckturm datasets from [15] and the Breisach dataset [40] for training and the datasets provided with COLMAP [33, 34] for testing.”
From Point Clouds to Mesh Using Regression (2017, background): “Similarly Ummenhofer et al. [30] proposed to globally optimize the signed distance ﬁeld on a hierarchical oct-tree.”
Deep blending for free-viewpoint image-based rendering (2018, methodology): “We also show InsideOut reinement with heuristic blending that is clearly worse; this is conirmed quantitatively in the leave-one-out rephotography [Waechter et al. 2017] results in Fig.”
Deep Local Shapes: Learning Local SDF Priors for Detailed 3D Reconstruction (2020, methodology): “urface patches [39,29,50]. Global approaches. Global implicit function approximation methods aim to approximate single continuous signed distance functions using, for example, kernel-based techniques [7,27,48,16]. Visibility or free space methods estimate which subset of 3D space is occupied, often ”