Matching with shape contexts
By Serge J. Belongie, Jitendra Malik. 2000 Proceedings Workshop on Content-based Access of Image and Video Libraries, 2000
339 citations recorded by the source corpus.
Subfields: Computer Vision and Pattern Recognition
View Paper: https://www.semanticscholar.org/paper/fe2d1a20c5a14653682ab903bf7727b0c37ddf0f
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
- Xiaogang Wang
- Luc Van Gool
- Jitendra Malik
- Dacheng Tao
- Cordelia Schmid
- Li Fei-Fei
- Trevor Darrell
- Ming-Hsuan Yang
Citations in the corpus
- On Aligning Curves (2003, methodology): “shape contexts [ 31 ] uses the global features and has problems differentiating between similar charact ers like “U” and “V”.”
- A Reflective Symmetry Descriptor (2002, methodology): “In this respect, our goals are similar to previously described shape descriptors [17,18,19,20], of which some recent examples include spin images [21], harmonic shape images [22], shape contexts[23,24], and Extended Gaussian Images [25].”
- Recognition of Shapes by Editing Shock Graphs (2001, methodology): “Shapes have been represented by their outline curves [1, 3, 14, 21], point sets, feature sets [2, 5], and by their medial axis [11, 13, 15, 18, 23], among others.”
- Recognition of shapes by editing their shock graphs (2004, background): “This similarity is not captured by shape context [4] since articulation (disproportionately) affects the global signature (shape context) of each point.”
- A survey on still image based human action recognition (2014, methodology): “Shape context (SC) was proposed by Belongie and Malik in 2000 [64] for shape feature extraction for object matching.”
- A survey of vision-based methods for action representation, segmentation and recognition (2011, methodology): “[18] proposes a view invariant pose representation based on a voxel reconstruction and cylindrical 3D histograms similar to the 2D shape context descriptor [3].”
- Hidden Part Models for Human Action Recognition: Probabilistic versus Max Margin (2011, background): “If an image is represented by a set of patches found by interest-point operators, one particular way to solve this problem is to first find the correspondence between patches in the two images, then learn a binary classifier based on the result of the correspondence [41].”