Probabilistic reasoning for assembly-based 3D modeling

By Siddhartha Chaudhuri, Evangelos Kalogerakis, Leonidas J. Guibas, Vladlen Koltun. 2011

282 citations recorded by the source corpus.

Subfields: Computational Mechanics, Computer Vision and Pattern Recognition, Control and Systems Engineering

How researchers see this paper

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

In Computational Mechanics

In Computer Vision and Pattern Recognition

In Control and Systems Engineering

Citations in the corpus

Learning a Probabilistic Latent Space of Object Shapes via 3D Generative-Adversarial Modeling (2016, background): “…in the graphics and vision community, and the relevant literature is very rich [Carlson, 1982, Tangelder and Veltkamp, 2008, Van Kaick et al., 2011, Blanz and Vetter, 1999, Kalogerakis et al., 2012, Chaudhuri et al., 2011, Xue et al., 2012, Kar et al., 2015, Bansal et al., 2016, Wu et al., 2016].”
VConv-DAE: Deep Volumetric Shape Learning Without Object Labels (2016, methodology): “Part and Symmetry based Shape Synthesis There exist some work [3,4] based on CAD models that use an assembly-based approach to build deformable partbased models.”
Synthesizing 3D Shapes via Modeling Multi-view Depth Maps and Silhouettes with Deep Generative Networks (2017, background): “Shape synthesis is then essentially to memorize and recombine these parts [10, 3].”
SurfNet: Generating 3D Shape Surfaces Using Deep Residual Networks (2017, background): “Recent approaches for assembly-based 3D shape creation from components use probabilistic models [5, 15], or deep-learned models [13].”
Data-driven structural priors for shape completion (2015, background): “estimate part compatibility [Chaudhuri et al. 2011; Kalogerakis et al. 2012; Xu et al. 2012], represent relations between parts [Fish et al.”
Garment modeling with a depth camera (2015, methodology): “We follow [Chaudhuri et al. 2011; Kalogerakis et al. 2012] to build a Bayesian network that captures the contextual prior from the labeled image database.”
Example-based synthesis of 3D object arrangements (2012, methodology): “Component-based object modeling Recent work has demonstrated a graphical model for individual objects (such as chairs, planes, and boats) that encodes the cardinality, style, and adjacencies of object sub-components [Kalogerakis et al. 2012; Chaudhuri et al. 2011].”
Deformation-driven shape correspondence via shape recognition (2017, methodology): “Parallel works explored the use of part-based Bayesian networks [Chaudhuri et al. 2011; Kalogerakis et al. 2012] and modular templates [Kim et al. 2013; Fish et al. 2014] to represent both continuous and discrete variations.”
A probabilistic model for component-based shape synthesis (2012, methodology): “The probabilistic model developed by Chaudhuri et al. [2011] can also be used to synthesize complete novel shapes, although it was not designed for this purpose and in our experiments generally produced shapes of low plausibility (Figure 9).”
ClipGen: A Deep Generative Model for Clipart Vectorization and Synthesis (2021, methodology): “Many assembly-based 3D modeling methods provide a probabilistic model to infer potential components for assembling under fully-automatic [4] and semi-automatic [21] settings.”
PolyGen: An Autoregressive Generative Model of 3D Meshes (2020, methodology): “Existing approaches to 3D object synthesis rely on the recombination and deformation of template models (Kaloger-akis et al., 2012; Chaudhuri et al., 2011), or a parametric shape family (Smelik et al., 2014).”
3 D ShapeNets : A Deep Representation for Volumetric Shape Modeling (2015, methodology): “While there is some good research on shape synthesis [7] [16] and shape reconstruction [27] [22], they are mostly limited to part-based assembly and suffer from establishing bad local correspondences.”
Joint shape segmentation with linear programming (2011, background): “Many shape processing and 3D modeling applications benefit from automatic segmentation of shapes into components that appear natural [Shamir 2008; Chen et al. 2009; Chaudhuri et al. 2011].”
PTR: A Benchmark for Part-based Conceptual, Relational, and Physical Reasoning (2021, background): “Restricted by the annotations in existing large-scale datasets, previous part-based 3D understanding works mainly focus on identifying parts [15, 30, 26, 55, 32, 58] and part relations [6, 10, 31, 35, 60, 50] for individual 3D objects.”
3D ShapeNets: A deep representation for volumetric shapes (2014, methodology): “While there is significant progress on shape synthesis [7, 17] and recovery [27], they are mostly limited to partbased assembly and heavily relies on expensive part annotation.”