COSMOS - A Representation Scheme for 3D Free-Form Objects

By C. Dorai, Anil K. Jain. 1997

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

VoxelNet: End-to-End Learning for Point Cloud Based 3D Object Detection (2017, methodology): “Some of the earlier methods for feature representation are [41, 8, 7, 19, 42, 35, 6, 27, 1, 36, 2, 25, 26].”
Matching 2.5D face scans to 3D models (2006, background): “Nine representative shapes on the shape index scale [30].”
Performance Evaluation of 3D Keypoint Detectors (2012, methodology): “It measures the saliency of a vertex according to its Shape Index (SI), as defined by Dorai and Jain (1997), which, in turn, is based on the maximum and minimum principal curvatures at the vertex,”
Using Spin Images for Efficient Object Recognition in Cluttered 3D Scenes (1999, background): “Viewercentered representations [6] describe surface data with respect to a coordinate system dependent on the view of the surface.”
Rapid object indexing using locality sensitive hashing and joint 3D-signature space estimation (2006, methodology): “LSH is a stochastic method based on the theory of random projections that was shown to be significantly faster compared with other fast nearest neighbor search methods, such as the dynamic space partitioning, and more robust to the choice of the internal parameter settings in higher dimensional…”
Topology matching for fully automatic similarity estimation of 3D shapes (2001, methodology): “Curvature is a possible candidate for this search key, and methods using curvature distribution have provided some good results for uses such as 3D shape pose estimation [11, 18, 22, 35, 39].”
Three-dimensional computer-aided diagnosis scheme for detection of colonic polyps (2001, background): “Every distinct shape, except for the plane, corresponds to a unique value of [27], [28].”
Scale-hierarchical 3D object recognition in cluttered scenes (2010, background): “Flynn and Jain [4] prune this space by using various unary and binary predicates for 3D recognition of objects with planar, cylindrical, and spherical surface types.”
Voting-based pose estimation for robotic assembly using a 3D sensor (2012, background): “Several feature descriptors and matching algorithms have been proposed [12], [13], [14], [15], [16].”
SHOT: Unique signatures of histograms for surface and texture description (2014, background): “Within the same subclass, Local Surface Patches [6] computes histograms of normals and shape indexes [26] of the points belonging to the support.”
3D Facial Expression Recognition Based on Primitive Surface Feature Distribution (2006, methodology): “For example, (1) the shape index based classification [7]; (2) Tanaka’s method [21], which categorized eight distinct features to describe local shapes according to the sign of two principal curvatures; (3) the topographic classification method [23, 26], which has been used for analyzing the 3D topo”
MVX-Net: Multimodal VoxelNet for 3D Object Detection (2019, background): “Earlier approaches ([6], [36], [1], [8], [37]) employ hand-crafted representations and achieve satisfactory results in the presence of rich and detailed 3D information.”
A Survey on Human Motion Analysis from Depth Data (2013, methodology): “[88] use curvature in conjunction with the Shape Index proposed by [89] to locate facial feature points from range scans of faces.”
On the Repeatability and Quality of Keypoints for Local Feature-based 3D Object Retrieval from Cluttered Scenes (2009, background): “Dorai and Jain [9] proposed a shape index using the principal curvatures of the surface for 3D free-form object recognition.”
3D free-form object recognition in range images using local surface patches (2004, methodology): “…of the shape of a surface at a point p, is defined by (1) where k 1 and k 2 are maximum and minimum principal curvatures, respectively S i ðpÞ ¼ 1 2 À 1 p tan À1 k 1 ðpÞ þ k 2 ðpÞ k 1 ðpÞ À k 2 ðpÞ : ð1Þ With this definition, all shapes are mapped into the interval [0] [1] (Dorai and Jain, 1997).”
A Survey Of Free-Form Object Representation and Recognition Techniques (2001, methodology): “Surface curvature has also been used to classify local surface shape into a small set of representative shapes [9, 31].”
Random Forests for Real Time 3D Face Analysis (2012, methodology): “Colbry et al. (2005) use curvature in conjunction with the Shape Index proposed by Dorai and Jain (1997) to locate facial feature points from range scans of faces.”
Three-Dimensional Model-Based Object Recognition and Segmentation in Cluttered Scenes (2006, background): “COSMOS [13] requires the calculation of principal curvatures which are not only sensitive to noise but also require the underlying surface to be smooth and twice differentiable.”