Revisiting Skeletons from Natural Images

Date

2015-01-01

Author

Erdem, Mehmet Erkut
Tarı, Zehra Sibel

Metadata

Show full item record

Item Usage Stats

75
views

0
downloads

In the last two decades there have been several works promoting shape fields that implicitly encode local convexity/concavity properties of the shape boundary. These shape fields are formulated either as solutions to Poisson type PDEs or via heuristic approximations to them. The v-field of Tari-Shah-Pien, can be computed directly from a real image; thus, suggests a mechanism to bridge low level visual processing and high level shape computations. We revisit Tari, Shah and Pien’s v-field approach and extend its application to complex images with texture. We relate v-field value at a skeleton point to the distance of the point from a putative shape boundary, and use this relation to extract semantic image patches. At the end of the chapter, we experimentally compare the medial locus computed from the new v-field to that of Kimia et al.

Subject Keywords

Shape Interior, Visual Processing, High Level, Edge Diffusivity, Skeleton Point, Shape Boundary

URI

http://www.springer.com/in/book/9783319163475
https://hdl.handle.net/11511/85467
https://link.springer.com/book/10.1007/978-3-319-16348-2

Relation

Research in Shape Modeling

Collections

Department of Computer Engineering, Book / Book chapter

Suggestions

OpenMETU
Core

Revisiting Skeletons from Natural Images Erdem, Erkut; Tarı, Zehra Sibel (2013-07-01) In the last two decades there have been several works promoting shape fields that implicitly encode local convexity/concavity properties of the shape boundary. These shape fields are formulated either as solutions to Poisson type PDEs or via heuristic approximations to them. The v-field of Tari-Shah-Pien, can be computed directly from a real image; thus, suggests a mechanism to bridge low level visual processing and high level shape computations. We revisit Tari, Shah and Pien's v-field approach and extend ...
A new approach to the classical Stokes flow problem .1. Methodology and first-order analytical results Taşeli, Hasan (1997-02-17) The problem of determining the axisymmetric Stokes flow past an arbitrary body, the boundary shape of which can be represented by an analytic function, is examined by developing an exact method. An appropriate nonorthogonal coordinate system is introduced, and it is shown that the Hilbert space to which the stream function belongs is spanned by the set of Gegenbauer polynomials based on the physical argument that the drag on a body should be finite. The partial differential equation of the original problem ...
A Numerically Stable Algorithm for Eccentrically Metamaterial Covered Circular Cylinders DİKMEN, FATİH; SEVER, EMRAH; TUCHKİN, YURY ALEXANDEROVİCH; Sabah, Cumali (2016-08-18) The regularization for monochromatic TM/TE-z polarized waves scattering from multiple non intersecting circular penetrable boundaries has recently been proven to be a requisite for its stable numerical implementation for a wide scope of parameters. The validity and necessity of corresponding regularization algorithm will be demonstrated for medium parameters which are from a double negative (DNG) material media. The preliminary numerical results which already prove the properties mentioned above are given i...
Revisiting Shape Embedding Keles, Hacer Yalim; Ozkar, Mine; Tarı, Zehra Sibel (2009-09-19) We propose and describe a working computer implementation for shape grammars that handles embedding relations in two dimensional shapes. The technical framework proposed explores a graph data structure to temporarily represent boundary elements of shapes and how they are assembled. With the associated algorithms, this structure enables a systematic search for parts. The employment of user defined constraints allows for an interactive search. In accordance with the continuous character of shapes, the study p...
A probabilistic sparse skeleton based object detection Altinoklu, Burak; Ulusoy, İlkay; Tarı, Zehra Sibel (Elsevier BV, 2016-11) We present a Markov Random Field (MRF) based skeleton model for object shape and employ it in a probabilistic chamfer-matching framework for shape based object detection. Given an object category, shape hypotheses are generated from a set of sparse (coarse) skeletons guided by suitably defined unary and binary potentials at and between shape parts. The Markov framework assures that the generated samples properly reflect the observed or desired shape variability. As the model employs a sparsely sampled skele...

Citation Formats

M. E. Erdem and Z. S. Tarı, Revisiting Skeletons from Natural Images. 2015, p. 112.