The Experts below are selected from a list of 282 Experts worldwide ranked by ideXlab platform
Stefano Soatto - One of the best experts on this subject based on the ideXlab platform.
-
one shot Integral Invariant shape priors for variational segmentation
Energy Minimization Methods in Computer Vision and Pattern Recognition, 2005Co-Authors: Siddharth Manay, Daniel Cremers, Anthony Yezzi, Stefano SoattoAbstract:We match shapes, even under severe deformations, via a smooth re-parametrization of their Integral Invariant signatures. These robust signatures and correspondences are the foundation of a shape energy functional for variational image segmentation. Integral Invariant shape templates do not require registration and allow for significant deformations of the contour, such as the articulation of the object's parts. This enables generalization to multiple instances of a shape from a single template, instead of requiring several templates for searching or training. This paper motivates and presents the energy functional, derives the gradient descent direction to optimize the functional, and demonstrates the method, coupled with a data term, on real image data where the object's parts are articulated.
-
EMMCVPR - One-Shot Integral Invariant shape priors for variational segmentation
Lecture Notes in Computer Science, 2005Co-Authors: Siddharth Manay, Daniel Cremers, Anthony Yezzi, Stefano SoattoAbstract:We match shapes, even under severe deformations, via a smooth re-parametrization of their Integral Invariant signatures. These robust signatures and correspondences are the foundation of a shape energy functional for variational image segmentation. Integral Invariant shape templates do not require registration and allow for significant deformations of the contour, such as the articulation of the object's parts. This enables generalization to multiple instances of a shape from a single template, instead of requiring several templates for searching or training. This paper motivates and presents the energy functional, derives the gradient descent direction to optimize the functional, and demonstrates the method, coupled with a data term, on real image data where the object's parts are articulated.
-
Integral Invariant signatures
European Conference on Computer Vision, 2004Co-Authors: Siddharth Manay, Anthony Yezzi, Byungwoo Hong, Stefano SoattoAbstract:For shapes represented as closed planar contours, we introduce a class of functionals that are Invariant with respect to the Euclidean and similarity group, obtained by performing Integral operations. While such Integral Invariants enjoy some of the desirable properties of their differential cousins, such as locality of computation (which allows matching under occlusions) and uniqueness of representation (in the limit), they are not as sensitive to noise in the data. We exploit the Integral Invariants to define a unique signature, from which the original shape can be reconstructed uniquely up to the symmetry group, and a notion of scale-space that allows analysis at multiple levels of resolution. The Invariant signature can be used as a basis to define various notions of distance between shapes, and we illustrate the potential of the Integral Invariant representation for shape matching on real and synthetic data.
-
ECCV (4) - Integral Invariant Signatures
Lecture Notes in Computer Science, 2004Co-Authors: Siddharth Manay, Anthony Yezzi, Byungwoo Hong, Stefano SoattoAbstract:For shapes represented as closed planar contours, we introduce a class of functionals that are Invariant with respect to the Euclidean and similarity group, obtained by performing Integral operations. While such Integral Invariants enjoy some of the desirable properties of their differential cousins, such as locality of computation (which allows matching under occlusions) and uniqueness of representation (in the limit), they are not as sensitive to noise in the data. We exploit the Integral Invariants to define a unique signature, from which the original shape can be reconstructed uniquely up to the symmetry group, and a notion of scale-space that allows analysis at multiple levels of resolution. The Invariant signature can be used as a basis to define various notions of distance between shapes, and we illustrate the potential of the Integral Invariant representation for shape matching on real and synthetic data.
Siddharth Manay - One of the best experts on this subject based on the ideXlab platform.
-
one shot Integral Invariant shape priors for variational segmentation
Energy Minimization Methods in Computer Vision and Pattern Recognition, 2005Co-Authors: Siddharth Manay, Daniel Cremers, Anthony Yezzi, Stefano SoattoAbstract:We match shapes, even under severe deformations, via a smooth re-parametrization of their Integral Invariant signatures. These robust signatures and correspondences are the foundation of a shape energy functional for variational image segmentation. Integral Invariant shape templates do not require registration and allow for significant deformations of the contour, such as the articulation of the object's parts. This enables generalization to multiple instances of a shape from a single template, instead of requiring several templates for searching or training. This paper motivates and presents the energy functional, derives the gradient descent direction to optimize the functional, and demonstrates the method, coupled with a data term, on real image data where the object's parts are articulated.
-
EMMCVPR - One-Shot Integral Invariant shape priors for variational segmentation
Lecture Notes in Computer Science, 2005Co-Authors: Siddharth Manay, Daniel Cremers, Anthony Yezzi, Stefano SoattoAbstract:We match shapes, even under severe deformations, via a smooth re-parametrization of their Integral Invariant signatures. These robust signatures and correspondences are the foundation of a shape energy functional for variational image segmentation. Integral Invariant shape templates do not require registration and allow for significant deformations of the contour, such as the articulation of the object's parts. This enables generalization to multiple instances of a shape from a single template, instead of requiring several templates for searching or training. This paper motivates and presents the energy functional, derives the gradient descent direction to optimize the functional, and demonstrates the method, coupled with a data term, on real image data where the object's parts are articulated.
-
Integral Invariant signatures
European Conference on Computer Vision, 2004Co-Authors: Siddharth Manay, Anthony Yezzi, Byungwoo Hong, Stefano SoattoAbstract:For shapes represented as closed planar contours, we introduce a class of functionals that are Invariant with respect to the Euclidean and similarity group, obtained by performing Integral operations. While such Integral Invariants enjoy some of the desirable properties of their differential cousins, such as locality of computation (which allows matching under occlusions) and uniqueness of representation (in the limit), they are not as sensitive to noise in the data. We exploit the Integral Invariants to define a unique signature, from which the original shape can be reconstructed uniquely up to the symmetry group, and a notion of scale-space that allows analysis at multiple levels of resolution. The Invariant signature can be used as a basis to define various notions of distance between shapes, and we illustrate the potential of the Integral Invariant representation for shape matching on real and synthetic data.
-
ECCV (4) - Integral Invariant Signatures
Lecture Notes in Computer Science, 2004Co-Authors: Siddharth Manay, Anthony Yezzi, Byungwoo Hong, Stefano SoattoAbstract:For shapes represented as closed planar contours, we introduce a class of functionals that are Invariant with respect to the Euclidean and similarity group, obtained by performing Integral operations. While such Integral Invariants enjoy some of the desirable properties of their differential cousins, such as locality of computation (which allows matching under occlusions) and uniqueness of representation (in the limit), they are not as sensitive to noise in the data. We exploit the Integral Invariants to define a unique signature, from which the original shape can be reconstructed uniquely up to the symmetry group, and a notion of scale-space that allows analysis at multiple levels of resolution. The Invariant signature can be used as a basis to define various notions of distance between shapes, and we illustrate the potential of the Integral Invariant representation for shape matching on real and synthetic data.
Anthony Yezzi - One of the best experts on this subject based on the ideXlab platform.
-
one shot Integral Invariant shape priors for variational segmentation
Energy Minimization Methods in Computer Vision and Pattern Recognition, 2005Co-Authors: Siddharth Manay, Daniel Cremers, Anthony Yezzi, Stefano SoattoAbstract:We match shapes, even under severe deformations, via a smooth re-parametrization of their Integral Invariant signatures. These robust signatures and correspondences are the foundation of a shape energy functional for variational image segmentation. Integral Invariant shape templates do not require registration and allow for significant deformations of the contour, such as the articulation of the object's parts. This enables generalization to multiple instances of a shape from a single template, instead of requiring several templates for searching or training. This paper motivates and presents the energy functional, derives the gradient descent direction to optimize the functional, and demonstrates the method, coupled with a data term, on real image data where the object's parts are articulated.
-
EMMCVPR - One-Shot Integral Invariant shape priors for variational segmentation
Lecture Notes in Computer Science, 2005Co-Authors: Siddharth Manay, Daniel Cremers, Anthony Yezzi, Stefano SoattoAbstract:We match shapes, even under severe deformations, via a smooth re-parametrization of their Integral Invariant signatures. These robust signatures and correspondences are the foundation of a shape energy functional for variational image segmentation. Integral Invariant shape templates do not require registration and allow for significant deformations of the contour, such as the articulation of the object's parts. This enables generalization to multiple instances of a shape from a single template, instead of requiring several templates for searching or training. This paper motivates and presents the energy functional, derives the gradient descent direction to optimize the functional, and demonstrates the method, coupled with a data term, on real image data where the object's parts are articulated.
-
Integral Invariant signatures
European Conference on Computer Vision, 2004Co-Authors: Siddharth Manay, Anthony Yezzi, Byungwoo Hong, Stefano SoattoAbstract:For shapes represented as closed planar contours, we introduce a class of functionals that are Invariant with respect to the Euclidean and similarity group, obtained by performing Integral operations. While such Integral Invariants enjoy some of the desirable properties of their differential cousins, such as locality of computation (which allows matching under occlusions) and uniqueness of representation (in the limit), they are not as sensitive to noise in the data. We exploit the Integral Invariants to define a unique signature, from which the original shape can be reconstructed uniquely up to the symmetry group, and a notion of scale-space that allows analysis at multiple levels of resolution. The Invariant signature can be used as a basis to define various notions of distance between shapes, and we illustrate the potential of the Integral Invariant representation for shape matching on real and synthetic data.
-
ECCV (4) - Integral Invariant Signatures
Lecture Notes in Computer Science, 2004Co-Authors: Siddharth Manay, Anthony Yezzi, Byungwoo Hong, Stefano SoattoAbstract:For shapes represented as closed planar contours, we introduce a class of functionals that are Invariant with respect to the Euclidean and similarity group, obtained by performing Integral operations. While such Integral Invariants enjoy some of the desirable properties of their differential cousins, such as locality of computation (which allows matching under occlusions) and uniqueness of representation (in the limit), they are not as sensitive to noise in the data. We exploit the Integral Invariants to define a unique signature, from which the original shape can be reconstructed uniquely up to the symmetry group, and a notion of scale-space that allows analysis at multiple levels of resolution. The Invariant signature can be used as a basis to define various notions of distance between shapes, and we illustrate the potential of the Integral Invariant representation for shape matching on real and synthetic data.
Daniel Cremers - One of the best experts on this subject based on the ideXlab platform.
-
one shot Integral Invariant shape priors for variational segmentation
Energy Minimization Methods in Computer Vision and Pattern Recognition, 2005Co-Authors: Siddharth Manay, Daniel Cremers, Anthony Yezzi, Stefano SoattoAbstract:We match shapes, even under severe deformations, via a smooth re-parametrization of their Integral Invariant signatures. These robust signatures and correspondences are the foundation of a shape energy functional for variational image segmentation. Integral Invariant shape templates do not require registration and allow for significant deformations of the contour, such as the articulation of the object's parts. This enables generalization to multiple instances of a shape from a single template, instead of requiring several templates for searching or training. This paper motivates and presents the energy functional, derives the gradient descent direction to optimize the functional, and demonstrates the method, coupled with a data term, on real image data where the object's parts are articulated.
-
EMMCVPR - One-Shot Integral Invariant shape priors for variational segmentation
Lecture Notes in Computer Science, 2005Co-Authors: Siddharth Manay, Daniel Cremers, Anthony Yezzi, Stefano SoattoAbstract:We match shapes, even under severe deformations, via a smooth re-parametrization of their Integral Invariant signatures. These robust signatures and correspondences are the foundation of a shape energy functional for variational image segmentation. Integral Invariant shape templates do not require registration and allow for significant deformations of the contour, such as the articulation of the object's parts. This enables generalization to multiple instances of a shape from a single template, instead of requiring several templates for searching or training. This paper motivates and presents the energy functional, derives the gradient descent direction to optimize the functional, and demonstrates the method, coupled with a data term, on real image data where the object's parts are articulated.
Li Zi-ping - One of the best experts on this subject based on the ideXlab platform.
-
Poincaré-Cartan Integral Invariant of a nonholonomic constrained generalized mechanical system
Journal of Jianghan University, 2005Co-Authors: Li Ai-min, Zhang Ying, Li Zi-pingAbstract:Based on generalized Apell-Четаев constrained conditions and taking into a ccount the inherent higher-order nonholonomic constraints,the generalized Poincare-Cartan Integral Invariant for a generalized mechanical system with higher- order subsidiary nonholonomic constraints is formulated. We can show that the ex istence of Poincare-Cartan Integral Invariant for such a system is equivalent to the generalized canonical equation of a nonholonomic constrained generalized mechanical system.
-
Quantal Poincaré-Cartan Integral Invariant for Field Theory
International Journal of Theoretical Physics, 2004Co-Authors: Zhang Ying, Li Zi-pingAbstract:On the basis of the phase-space generating function of Green function for a system with a regular/singular Lagrangian, the quantal Poincare-Cartan Integral Invariant (PCII) for field theory is derived. This PCII is equivalent to the quantal canonical equations. For this case in which the Jacobian of the transformation does not equalto unity, the quantal PCII can still be derived. This case is different from the quantal first Noether theorem. The quantal PCII connected with canonical equations and canonical transformation is also discussed.
-
Generalized Canonical Noether Theorem and Poincaré–Cartan Integral Invariant for a System with a Singular High-Order Lagrangian and an Application
Communications in Theoretical Physics, 2001Co-Authors: Li Zi-ping, Li Rui-jieAbstract:Based on the canonical action, a generalized canonical first Noether theorem and Poicare–Cartan Integral-Invariant for a system with a singular high-order Lagrangian are derived. It is worth while to point out that the constraints are Invariant under the total variation of canonical variables including time. We can also deduce the result, which differs from the previous work to require that the constraints are Invariant under the simultaneous variations of canonical variables. A counter example to a conjecture of the Dirac for a system with a singular high-order Lagrangian is given, in which there is no linearization of constraint.
-
Generalized Noether Theorem and Poincaré-Cartan Integral Invariant for Singular High-order Lagrangian in Fields Theories
Science in China Series A-Mathematics Physics Astronomy & Technological Science, 1993Co-Authors: Li Zi-pingAbstract:A generalized first Noether theorem (GFNT) originating from the invariance under the finite continuous group for singular high-order Lagrangian and a generalized second Noether theorem (or generalized Noether identities (GNI)) for variant system under the infinite continuous group of field theory in canonical formalism are derived. The strong and weak conservation laws in canonical formalism are also obtained. It is pointed out that some variant systems also have Dirac constraint. Based on the canonical action, the generalized Poincare-Cartan Integral Invariant (GPCⅡ) for singular high-order Lagrangian in the field theory is deduced. Some confusions in literafure are clarified. The GPCⅡ connected with canonical equations and canonical transformation are discussed.