Intrinsic shape context descriptors for deformable shapes

By Iasonas Kokkinos, Michael M. Bronstein, Roee Litman, A. Bronstein. 2012

179 citations recorded by the source corpus.

Subfields: Computational Mechanics, 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 Vision and Pattern Recognition

Citations in the corpus

Geometric Deep Learning on Graphs and Manifolds Using Mixture Model CNNs (2016, background): “[26] introduced a generalization of CNNs on 2-dimensional manifolds, based on the definition of a local charting procedure in geodesic polar coordinates [22].”
Volumetric and Multi-view CNNs for Object Classification on 3D Data (2016, methodology): “Alternatives include models based on distances, angles, triangle areas, or tetrahedra volumes [23], local shape diameters measured at densely-sampled surface points [3], Heat kernel signatures [1, 17], or extensions of SIFT and SURF feature descriptors to 3D voxel grids [16].”
Multi-view Convolutional Neural Networks for 3D Shape Recognition (2015, background): “For example, shapes can be represented with histograms or bag-of-features models constructed out of surface normals and curvatures [15], distances, angles, triangle areas or tetrahedra volumes gathered at sampled surface points [25], properties of spherical functions defined in volumetric grids [16]”
SplineCNN: Fast Geometric Deep Learning with Continuous B-Spline Kernels (2017, methodology): “Note that we used neighborhoods of size 5 × 5 from the grid graph, to mirror the LeNet5 architecture with its 5× 5 filters.”
Learning shape correspondence with anisotropic convolutional neural networks (2016, methodology): “While relatively insensitive to triangulation [12], it may fail if the mesh is very irregular.”
FusionNet: 3D Object Classification Using Multiple Data Representations (2016, neutral): “Other shape descriptors include the Light Field Descriptor [3], Heat kernel signatures [2] [12] and SPH [11].”
MeshCNN: a network with an edge (2019, methodology): “[2015] introduced deep learning of local features on meshes (intrinsic mesh descriptors similar to the ones in [Kokkinos et al. 2012]), and has shown how to use these learned features for correspondence and retrieval.”
GVCNN: Group-View Convolutional Neural Networks for 3D Shape Recognition (2018, methodology): “The shape-based descriptor was designed based on the native 3D representations of objects, such as voxel grid [26], polygon mesh [2, 10], local shape diameters measured at densely sampled surface points [3], or extensions of the SIFT and SURF descriptors to 3D voxel grids [13].”
Geom-GCN: Geometric Graph Convolutional Networks (2020, background): “One can design a more sophisticated operator τ , such as borrowing the structure of descriptors in manifold geometry (Kokkinos et al., 2012; Monti et al., 2017), thereby preserving more and richer structural information in neighborhood.”
Geodesic Convolutional Neural Networks on Riemannian Manifolds (2015, methodology): “The success of image descriptors such as SIFT [31], HOG [13], MSER [33], and shape context [2] has led to several generalizations thereof to non- Euclidean domains (see e.g. [49] [14] [24], respectively).”
MeshNet: Mesh Neural Network for 3D Shape Representation (2018, background): “( Zaharescu et al. 2009) Intrinsic shape context (ISC) descriptor (Kokkinos et al. 2012) develops a generalization to surfaces and solves the problem of orientational ambiguity.”
Body Surface Context: A New Robust Feature for Action Recognition From Depth Videos (2014, background): “2-D shape descriptors, such as the shape context [35], were also extended to 3-D domain [36], [37], [41].”
Rotational Projection Statistics for 3D Local Surface Description and Object Recognition (2013, background): “Kokkinos et al. (2012) proposed a generalization of 2D shape context feature (Belongie et al., 2002) to curved surfaces, namely Intrinsic Shape Context (ISC).”
3D Object Recognition in Cluttered Scenes with Local Surface Features: A Survey (2014, background): “Another generalization of the 2D shape context is the “intrinsic shape context (ISC)” descriptor [116], which is invariant to isomet-”
Anisotropic Diffusion Descriptors (2016, methodology): “A similar ShapeNet model [MBBV15] used local geodesic polar coordinates [KBLB12] to deﬁne convo-lutions on meshes.”
3D object classification and retrieval with Spherical CNNs (2017, background): “The Intrinsic Shape Context descriptor (ISC) [18] is an extension of the 2D Shape Context [4] descriptor to triangular meshes.”