Deformation-driven shape correspondence via shape recognition

By Jun Li, Kai Xu, Siddhartha Chaudhuri, Ersin Yumer, Hao Zhang, Leonidas J. Guibas. 2017

176 citations recorded by the source corpus.

Many approaches to shape comparison and recognition start by establishing a shape correspondence. We "turn the table" and show that quality shape correspondences can be obtained by performing many shape recognition tasks. What is more, the method we develop computes a fine-grained, topology-varying part correspondence between two 3D shapes where the core evaluation mechanism only recognizes shapes globally. This is made possible by casting the part correspondence problem in a deformation-driven framework and relying on a data-driven "deformation energy" which rates visual similarity between deformed shapes and models from a shape repository. Our basic premise is that if a correspondence between two chairs (or airplanes, bicycles, etc.) is correct, then a reasonable deformation between the two chairs anchored on the correspondence ought to produce plausible, "chair-like" in-between shapes. Given two 3D shapes belonging to the same category, we perform a top-down, hierarchical search for part correspondences. For a candidate correspondence at each level of the search hierarchy, we deform one input shape into the other, while respecting the correspondence, and rate the correspondence based on how well the resulting deformed shapes resemble other shapes from ShapeNet belonging to the same category as the inputs. The resemblance, i.e., plausibility, is measured by comparing multi-view depth images over category-specific features learned for the various shape categories. We demonstrate clear improvements over state-of-the-art approaches through tests covering extensive sets of man-made models with rich geometric and topological variations.

Subfields: Computational Mechanics, Computer Graphics and Computer-Aided Design, Computer Vision and Pattern Recognition

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 Graphics and Computer-Aided Design

In Computer Vision and Pattern Recognition

Citations in the corpus

Learning Implicit Fields for Generative Shape Modeling (2018, background): “There have been a variety of 3D shape representations for deep learning of shapes, such as voxel grids [9, 14, 28, 45, 46], octrees [18, 35, 40, 43, 44], multi-view images [29, 39], point clouds [1, 12, 13, 32, 33, 47, 48], geometry images [37, 38], deformable mesh/patches [16, 38, 42, 47], and part”
Neural 3D Mesh Renderer (2017, background): “Primitive-based representations, which represent 3D objects using a set of primitives, such as cuboids, have also been investigated [14, 39].”
Learning Gradient Fields for Shape Generation (2020, neutral): “Common representations emerged for deep generative 3D modeling include voxel-based [23,63], mesh-based [3,48,20,27,35,57], and assembly-based techniques [34,41].”
BSP-Net: Generating Compact Meshes via Binary Space Partitioning (2019, methodology): “There have been recent works on learning structured 3D models, in particular, linear [58] or hierarchical [23, 57, 29, 31] organization of part bounding boxes.”
Deep Generative Models in Engineering Design: A Review (2021, background): “For this reason, we do not discuss these methods in detail and opt to list a few of the many noteworthy papers in this field from 2016 to 2021 for the reader to further explore at their discretion: [140, 141, 142, 143, 144, 29].”
SCORES (2018, methodology): “GRASS is designed to synthesize 3D shapes globally resembling exemplars.”
PartNet: A Large-Scale Benchmark for Fine-Grained and Hierarchical Part-Level 3D Object Understanding (2018, background): “Teaching machines to analyze parts is thus essential for many vision, graphics, and robotics applications, such as predicting object functionality [13, 14], human-object interactions [18], shape editing [28, 16], and shape generation [25, 41].”
MeshCNN: a network with an edge (2019, methodology): “Following their success in the image domain, extensions to the 3D domain were quickly made, with shapes represented and processed predominately in voxel [Li et al. 2017; Wu et al. 2015] or point cloud form [Qi et al. 2017b].”