Shape descriptors for non-rigid shapes with a single closed contour

By Longin Jan Latecki, Rolf Lakämper, U. Eckhardt. Proceedings IEEE Conference on Computer Vision and Pattern Recognition. CVPR 2000 (Cat. No.PR00662), 2000

940 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/3762424071e0cd029ddd1c489215a75e323134f1

How researchers see this paper

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

In Computer Vision and Pattern Recognition

Citations in the corpus

Correlating fourier descriptors of local patches for road sign recognition (2011, methodology): “Nowadays, there are more robust (with respect to global occlusion and non-rigid deformations) but also computationally more expensive methods for comparing the global outlines of objects [5, 6, 7].”
Recognition of shapes by editing their shock graphs (2004, methodology): “Left: The database of 216 shapes selected from the MPEG-7 test database [24].”
Analysis of planar shapes using geodesic paths on shape spaces (2004, background): “Although image understanding forms an important application of shape analysis, a theory should be more general in its scope.”
Hierarchical Matching of Deformable Shapes (2007, methodology): “The MPEG-7 shape database [15] is a widely used dataset for testing shape recognition methods.”
Classification of time series by shapelet transformation (2013, methodology): “It is used for testing contour/image and skeleton-based descriptors Latecki et al. (2000).”
An Efficient Earth Mover's Distance Algorithm for Robust Histogram Comparison (2007, methodology): “The MPEG7 CE-Shape-1 database [23] has been widely used to test various shape matching algorithms.”
Shape retrieval using triangle-area representation and dynamic space warping (2007, methodology): “The first database is the well-known MPEG-7 CE-Shape-1 database [12] which consists of 1400 images semantically classified into 70 classes.”
On Aligning Curves (2003, methodology): “The retrieval rate for our approach then is . The currently published best performing approaches use curvature scale space [35], corresponde nce of visual parts [ 34 ], and matching shape contexts [36], with retrieval rates of , , and , respectively.”
Under Consideration for Publication in Knowledge and Information Systems Improving Svm Classification on Imbalanced Time Series Data Sets with Ghost Points (neutral): “To show that they will work with different distance measures, we use both DTW and OSB as distance measures on the UCR Time Series data sets [15] and on the MPEG-7 Core Experiment CE-Shape-1 data set [18] for our experiments.”
Shape Matching and Object Recognition Using Shape Contexts (2002, methodology): “A comprehensive comparison of di erent shape descriptors for comparing silhouettes was done as part of the MPEG-7 standard activity [33], with the leading approaches being those due to Latecki et al.”
Shape Similarity Measure Based on Correspondence of Visual Parts (2000, background): “MPEG-7 Core Experiment CE-Shape-1 for shape desciptors. We report on this in [ 9 ].”
Learning Context-Sensitive Shape Similarity by Graph Transduction (2010, methodology): “We obtained the retrieval rate of 91.61 percent on part B of the MPEG-7 Core Experiment CE-Shape-1 data set [18], which is the highest ever bulls-eye score reported in the literature.”
Matching with shape contexts (2000, neutral): “[25], with a retrieval rate of 76.”
Recognition of Shapes by Editing Shock Graphs (2001, methodology): “We constructed a second database from samples of a very large database of shapes created for testing the compression rates for MPEG7, kindly provided by Latecki and Lakamper [10].”
Spectral Embedded Clustering: A Framework for In-Sample and Out-of-Sample Spectral Clustering (2011, methodology): “Eight high-dimensional datasets are used in the experiments, including three face datasets (AR [46], YALE-B [47], and CMU PIE [48]), two shape image datasets (MPEG7 [ 49 ] and COIL20 [50]) and three handwritten digit datasets (Optdigits [51], USPS [52], and MNIST [53]).”
A Bayesian network framework for relational shape matching (2003, background): “Most of the feature-based recognition approaches begin with the fundamental assumption that the template and/or the data consists of a single closed curve or surface ([15, 22, 25, 26, 13, 12]).”
Concept-Based Video Retrieval (2009, background): “A distinction can be made between shape features that describe a region or features describing the shape contour [123] [250].”