Example-based 3D scan completion

By Mark Pauly, Niloy J. Mitra, Joachim Giesen, Markus H. Gross, Leonidas J. Guibas. 2005

340 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.

Citations in the corpus

Shape Completion Using 3D-Encoder-Predictor CNNs and Shape Synthesis (2016, neutral): “For instance Pauly et al. [29] complete 3D scans by ﬁrst retrieving candidate models from a database, then perform a non-rigid mesh alignment and blend the results with the input scan.”
Local Implicit Grid Representations for 3D Scenes (2020, unknown): “[30] combines shape patches extracted directly from a set of examples, which limits the shape ex-pressibility.”
Morphing and Sampling Network for Dense Point Cloud Completion (2019, unknown): “Example-based approaches (Pauly et al. 2005; Sung et al. 2015; Shen et al. 2012) ﬁrst retrieve some similar models in a large shape database, and then deform and assemble the retrieved models to complete the partial shape.”
Trajectory-Based Skill Learning Using Generalized Cylinders (2018, unknown): “Nonrigid registration techniques have been widely used in medical imaging (Maintz and Viergever, 1998), computer vision (Belongie et al., 2002), and 3D modeling communities (Pauly et al., 2005).”
Object detection and classification from large‐scale cluttered indoor scans (2014, unknown): “We experimented with the method of Pauly et al. [PMG ∗ 05], but our setting is more challenging due to the presence of noise in the data and possible outliers in each cluster.”
High-Resolution Shape Completion Using Deep Neural Networks for Global Structure and Local Geometry Inference (2017, unknown): “One simple approach is to retrieve the most similar shape from a database and use it as a template that can be deformed to ﬁt the input raw data ( [19] [21]).”
ScanComplete: Large-Scale Scene Completion and Semantic Segmentation for 3D Scans (2017, unknown): “, by nonrigid registration such that they better fit the scan [25, 31].”
Möbius voting for surface correspondence (2009, unknown): “Some previous work has relied upon iterative optimization of nearly rigid deformations [Allen et al. 2003; Pauly et al. 2005; Mitra et al. 2007; Ovsjanikov et al. 2008], while others have explicitly defined error functions that measure global extrinsic deformations [Botsch and Sorkine 2008].”
Semantic Multi-view Stereo: Jointly Estimating Objects and Voxels (2017, background): “match a database of object shapes against an incomplete point cloud from a 3D scanner and then align the best fitting shape to reconstruct occluded regions [26].”
Learning 3D Shape Completion from Laser Scan Data with Weak Supervision (2018, unknown): “[33] pose shape completion as retrieval and alignment problem.”
A Field Model for Repairing 3D Shapes (2016, unknown): “The object models can be predefined using CAD model databases, e.g. [26, 10, 31].”
Learning from Demonstrations Through the Use of Non-rigid Registration (2013, unknown): “In 3D modeling, it has been used to fill in missing parts of scans [2].”
Blended intrinsic maps (2011, methodology): “This method is most commonly used for aligning surfaces related by a rigid transformation [Besl and McKay 1992], but has also been used for moderate non-rigid deformations [Allen et al. 2003; Brown and Rusinkiewicz 2007; Li et al. 2008; Pauly et al. 2005; Tevs et al. 2009; Ghosh et al. 2009].”
A Survey of Surface Reconstruction from Point Clouds (2017, unknown): “A database of objects may be used to best match the incomplete point cloud to a complete model [PMG ∗ 05], or a composition of parts from different models [SFCH12].”
PCN: Point Completion Network (2018, unknown): “Some [15, 24, 32, 34, 43] retrieve the complete shape directly while some [17, 19, 28, 45, 52] retrieve object parts and then assemble them to obtain the complete shape.”