Shape Matching and Object Recognition Using Shape Contexts

By Serge J. Belongie, Jitendra Malik, Jan Puzicha. 2002

6,908 citations recorded by the source corpus.

Subfields: 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 Computer Vision and Pattern Recognition

Citations in the corpus

Ensemble of exemplar-SVMs for object detection and beyond (2011, background): “Because most detections show good alignment to their associated exemplar, it is possible to transfer any available exemplar meta-data (segmentation, geometric structure, 3D model, etc.) directly onto the detections, which can then be used as part of overall scene understanding.”
Character Recognition in Natural Images (2009, background): “Shape Contexts(SC) (Belongie et al., 2002) is a descriptor for point sets and binary images.”
SIFT Flow: Dense Correspondence across Scenes and Its Applications (2011, methodology): “Jenny Yuen and Antonio Torralba are with the Computer Science and Artificial Intelligence Laboratory, Massachusetts Institute of Technology, Cambridge, MA 02139.”
Data-Driven Grasp Synthesis—A Survey (2013, background): “It is represented relative to a potential grasping point by shape contexts as introduced by Belongie et al. [64].”
Res2Net: A New Multi-Scale Backbone Architecture (2019, methodology): “Unsurprisingly, multi-scale features have been widely used in both conventional feature design [1], [31] and deep learning [6], [18], [29], [39], [44], [52].”
Structured Prediction, Dual Extragradient and Bregman Projections (2006, background): “Real-life examples of such problems include bipartite matchings in alignment of 2D shapes (Belongie et al., 2002) and word alignment of sentences from a source language to a target language in machine translation (Matusov et al.”
Expandable Data-Driven Graphical Modeling of Human Actions Based on Salient Postures (2008, methodology): “They include binary masks [28], moments [23], Fourier shape descriptors [16], Kendall’s shape description [29], and shape-context [30].”
Robust hand gesture recognition based on finger-earth mover's distance with a commodity depth camera (2011, background): “Categories and Subject Descriptors H.1.2 [User/Machine Systems]: Human information processing; I.4.8 [Scene Analysis]: Depth cues, Shape General Terms Algorithm Keywords Hand Gesture Recognition, Human-Computer Interaction, Kinect Sensor, Finger-Earth Mover’s Distance”
Unsupervised co-segmentation for 3D shapes using iterative multi-label optimization (2013, background): “Observing the study on feature selection in [2], we chose four shape descriptors in this paper, average geodesic distance (AGD), conformal factor (CF), shape diameter function (SDF) and shape contexts (SC) [30].”
Shape and Appearance Context Modeling (2007, methodology): “We propose to compute the shape labeled image with a procedure that is inspired by the idea of shape context [1, 16].”
A Bayesian network framework for relational shape matching (2003, background): “The spin image is a novel shape attribute at each point on the template surface as is shape context [2].”
Three-dimensional shape searching: state-of-the-art review and future trends (2005, neutral): “Some commonly used methods of 2D shape matching are through the use of moments [29], Fourier coefficients [30], curvature scale space [31], shape contexts [32], and skeletons [33].”
Plant species identification using digital morphometrics: A review (2012, methodology): “Shape similarity is measured using the InnerDistance Shape Context algorithm, which extends the shape context work of Belongie, Malik, and Puzicha (2002).”
Capturing Hands in Action Using Discriminative Salient Points and Physics Simulation (2015, methodology): “We use γa = 0.0015 Call, where Call is the total number of correspondences.”
Robust Video Surveillance for Fall Detection Based on Human Shape Deformation (2011, methodology): “The moving edge points extracted from two consecutive images are then matched using shape context [20].”
A sparse texture representation using local affine regions (2005, methodology): “Recent experimental data tends to support this intuition: SIFT descriptors have achieved better repeatability rates than filter banks and differential invariants in the comparative evaluations of Mikolajczyk and Schmid [34], while shape contexts have performed remarkably well for the application of ”
Word image matching using dynamic time warping (2003, methodology): “SC [1]: Shape context matching (see section 1.”