Marcelo Romero | CICYTTP-CONICET - Academia.edu (original) (raw)

Papers by Marcelo Romero

A comparison of several approaches that use graph matching and cascade filtering for landmark loc... more A comparison of several approaches that use graph matching and cascade filtering for landmark localization in 3D face data is presented. For the first method, we apply the structural graph matching algorithm ldquorelaxation by eliminationrdquo using a simple ldquodistance to local planerdquo node property and a ldquoEuclidean distancerdquo arc property. After the graph matching process has eliminated unlikely candidates, the most likely triplet is selected, by exhaustive search, as the minimum Mahalanobis distance over a six dimensional space, corresponding to three node variables and three arc variables. A second method uses state-of-the-art pose-invariant feature descriptors embedded into a cascade filter to localize the nose tip. After that, local graph matching is applied to localize the inner eye corners. We evaluate our systems by computing root mean square errors of estimated landmark locations against ground truth landmark localizations within the 3D Face Recognition Grand Challenge database. Our best system, which uses a novel pose-invariant shape descriptor, scores 99.77% successful localization of the nose and 96.82% successful localization of the eyes.

... Marcelo Romero and Nick Pears Department of Computer Science The University of York York, UK ... more ... Marcelo Romero and Nick Pears Department of Computer Science The University of York York, UK {mromero, nep}@cs.york.ac.uk ... of descriptor are introduced, the first is the point-pair spin image, which is related to the classical spin image of Johnson and Hebert, and the ...

International Journal of Computer Vision, 2010

We consider the problem of generating either pairwise-aligned or pose-normalised depth maps from ... more We consider the problem of generating either pairwise-aligned or pose-normalised depth maps from noisy 3D point clouds in a relatively unrestricted poses. Our system is deployed in a 3D face alignment application and consists of the following four stages: (i) data filtering, (ii) nose tip identification and sub-vertex localisation, (iii) computation of the (relative) face orientation, (iv) generation of either a pose aligned or a pose normalised depth map. We generate an implicit radial basis function (RBF) model of the facial surface and this is employed within all four stages of the process. For example, in stage (ii), construction of novel invariant features is based on sampling this RBF over a set of concentric spheres to give a spherically-sampled RBF (SSR) shape histogram. In stage (iii), a second novel descriptor, called an isoradius contour curvature signal, is defined, which allows rotational alignment to be determined using a simple process of 1D correlation. We test our system on both the University of York (UoY) 3D face dataset and the Face Recognition Grand Challenge (FRGC) 3D data. For the more challenging UoY data, our SSR descriptors significantly outperform three variants of spin images, successfully identifying nose vertices at a rate of 99.6%. Nose localisation performance on the higher quality FRGC data, which has only small pose variations, is 99.9%. Our best system successfully normalises the pose of 3D faces at rates of 99.1% (UoY data) and 99.6% (FRGC data).

An investigation to localise facial landmarks from 3D images is presented, without using any assu... more An investigation to localise facial landmarks from 3D images is presented, without using any assumption concerning facial pose. This paper introduces new surface descriptors, which are derived from either unstructured face data, or a radial basis function (RBF) model of the facial surface. Two new variants of feature descriptors are described, generally named as point - triplet descriptors because they require three vertices to be computed. The first is related to the classical depth map feature, which is referred to as weighted - interpolated depth map. The second variant of descriptors are derived from an implicit RBF model, they are referred to as surface RBF signature (SRS) features. Both variants of descriptors are able to encode surface information within a triangular region defined by a point - triplet into a surface signature, which could be useful not only for 3D face processing but also within a number of graph based retrieval applications. These descriptors are embedded into a system designed to localise the nose - tip and two inner - eye corners. Landmark localisation performance is reported by computing errors of estimated landmark locations against our respective ground --truth data from the Face Recognition Grand Challenge (FRGC) database.

A comparison of several approaches that use graph matching and cascade filtering for landmark loc... more A comparison of several approaches that use graph matching and cascade filtering for landmark localization in 3D face data is presented. For the first method, we apply the structural graph matching algorithm ldquorelaxation by eliminationrdquo using a simple ldquodistance to local planerdquo node property and a ldquoEuclidean distancerdquo arc property. After the graph matching process has eliminated unlikely candidates, the most likely triplet is selected, by exhaustive search, as the minimum Mahalanobis distance over a six dimensional space, corresponding to three node variables and three arc variables. A second method uses state-of-the-art pose-invariant feature descriptors embedded into a cascade filter to localize the nose tip. After that, local graph matching is applied to localize the inner eye corners. We evaluate our systems by computing root mean square errors of estimated landmark locations against ground truth landmark localizations within the 3D Face Recognition Grand Challenge database. Our best system, which uses a novel pose-invariant shape descriptor, scores 99.77% successful localization of the nose and 96.82% successful localization of the eyes.

... Marcelo Romero and Nick Pears Department of Computer Science The University of York York, UK ... more ... Marcelo Romero and Nick Pears Department of Computer Science The University of York York, UK {mromero, nep}@cs.york.ac.uk ... of descriptor are introduced, the first is the point-pair spin image, which is related to the classical spin image of Johnson and Hebert, and the ...

International Journal of Computer Vision, 2010

We consider the problem of generating either pairwise-aligned or pose-normalised depth maps from ... more We consider the problem of generating either pairwise-aligned or pose-normalised depth maps from noisy 3D point clouds in a relatively unrestricted poses. Our system is deployed in a 3D face alignment application and consists of the following four stages: (i) data filtering, (ii) nose tip identification and sub-vertex localisation, (iii) computation of the (relative) face orientation, (iv) generation of either a pose aligned or a pose normalised depth map. We generate an implicit radial basis function (RBF) model of the facial surface and this is employed within all four stages of the process. For example, in stage (ii), construction of novel invariant features is based on sampling this RBF over a set of concentric spheres to give a spherically-sampled RBF (SSR) shape histogram. In stage (iii), a second novel descriptor, called an isoradius contour curvature signal, is defined, which allows rotational alignment to be determined using a simple process of 1D correlation. We test our system on both the University of York (UoY) 3D face dataset and the Face Recognition Grand Challenge (FRGC) 3D data. For the more challenging UoY data, our SSR descriptors significantly outperform three variants of spin images, successfully identifying nose vertices at a rate of 99.6%. Nose localisation performance on the higher quality FRGC data, which has only small pose variations, is 99.9%. Our best system successfully normalises the pose of 3D faces at rates of 99.1% (UoY data) and 99.6% (FRGC data).

An investigation to localise facial landmarks from 3D images is presented, without using any assu... more An investigation to localise facial landmarks from 3D images is presented, without using any assumption concerning facial pose. This paper introduces new surface descriptors, which are derived from either unstructured face data, or a radial basis function (RBF) model of the facial surface. Two new variants of feature descriptors are described, generally named as point - triplet descriptors because they require three vertices to be computed. The first is related to the classical depth map feature, which is referred to as weighted - interpolated depth map. The second variant of descriptors are derived from an implicit RBF model, they are referred to as surface RBF signature (SRS) features. Both variants of descriptors are able to encode surface information within a triangular region defined by a point - triplet into a surface signature, which could be useful not only for 3D face processing but also within a number of graph based retrieval applications. These descriptors are embedded into a system designed to localise the nose - tip and two inner - eye corners. Landmark localisation performance is reported by computing errors of estimated landmark locations against our respective ground --truth data from the Face Recognition Grand Challenge (FRGC) database.