How effective are landmarks and their geometry for face recognition?

Shi, Junbo; Samal, Ashok; Marx, David B.

doi:10.1016/j.cviu.2005.10.002

Cited by 121 publications

(82 citation statements)

References 30 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Perhaps our work is more closely related to earlier research by Shi et al [23,24] on face recognition using geometric features, where they used ratio features computed from a few anatomical landmarks. However, we take a more comprehensive look at the explicit use of facial geometry in solving the problem of gender classification.…”

Section: Related Workmentioning

confidence: 70%

“…We cannot directly use the landmark coordinates, since they will be sensitive to translation, scaling, and 2D rotation of face images. One could consider all distance ratios defined by sets of four landmarks, or triangular features defined by any three non-collinear landmarks [24]. The problem here is computational complexity.…”

Section: Metrological Featuresmentioning

confidence: 99%

“…Facial landmarks can be divided into three broad categories [24]: anatomical landmarks, mathematical landmarks, and pseudo-landmarks. Anatomical landmarks are biologically meaningful points defined as standard reference points on the face and head, such as pupil, dacryon, nasion or pogonion, and they are often used by scientists and physicians.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Can facial metrology predict gender?

Cao

Chen

Piccirilli

et al. 2011

2011 International Joint Conference on Biometrics (IJCB)

View full text Add to dashboard Cite

We investigate the question of whether facial metrology can be exploited for reliable gender prediction. A new method based solely on metrological information from facial landmarks is developed. Here, metrological features are defined in terms of specially normalized angle and distance measures and computed based on given landmarks on facial images. The performance of the proposed metrologybased method is compared with that of a state-of-the-art appearance-based method for gender classification. Results are reported on two standard face databases, namely, MUCT and XM2VTS containing 276 and 295 images, respectively. The performance of the metrology-based approach was slightly lower than that of the appearancebased method by only about 3.8% for the MUCT database and about 5.7% for the XM2VTS database.

show abstract

Section: Related Workmentioning

confidence: 70%

Section: Metrological Featuresmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Can facial metrology predict gender?

Cao

Chen

Piccirilli

et al. 2011

2011 International Joint Conference on Biometrics (IJCB)

View full text Add to dashboard Cite

show abstract

“…The problem is that, at least for landmark analysis, they rely on homologous points. For example in face recognition (Shi et al 2006), the tip of a nose may be considered homologous (in the sense of evolutionary origins, growth and development etc.) as that of another human, however the further we move away from the same species or taxon the more difficult it becomes to place the landmark (e.g.…”

mentioning

confidence: 99%

Species differentiation of slipper orchids using color image analysis

2013

View full text Add to dashboard Cite

abstraCt. A number of automated species recognition systems have been developed recently to aid nonprofessionals in the identification of taxa. These systems have primarily used geometric morphometric based techniques, however issues surround their wider applicability due to the need for homologous landmarks. Here we investigate the use of color to discriminate species using the two horticulturally important slipper orchid genera of Paphiopedilum and Phragmipedium as model systems. The ability to differentiate the various taxonomic groups varied, depending on the size of the group, diversity of colors within the group, and the background of the image. In this study the image analysis was conducted with images of single flowers of the species, however since flowers are ephemeral, flowering for a relatively short period of time, such analysis should be extended to vegetative parts, particularly as this is the form in which they are most often traded internationally.rEsumEn. Una gran cantidad de sistemas de reconocimiento automático de especies se han desarrollado en los últimos años, como ayuda a aquellas personas que no son especialistas en la identificación de especies. Estos sistemas han utilizado sistemas de reconocimiento automático basados en geometría morfométrica, sin embargo existen límites debido a la necesidad de encontrar puntos de georreferenciación en los diferentes organismos. En este artículo investigamos el uso de los colores para diferenciar especies en los géneros Paphiopedilum y Phragmipedium, ambos con gran importancia en la horticultura. La capacidad de discriminación varía entre los grupos taxonómicos, dependiendo del tamaño del taxón, la variedad de colores entre las especies y el fondo de las imágenes. En este estudio el análisis de imágenes se ha llevado a cabo con fotografías de flores individuales. No obstante dado que las flores son órganos efímeros, en el futuro esta investigación incluirá partes vegetativas, ya que es en estado vegetativo la forma en la que se suele comerciar internacionalmente más a menudo.

show abstract

“…For instance, Shi et. al [7] has recently proposed a refined Procrustes distance based on principal component analysis. The configurations (the landmark representations of the faces) are first centered at the origin and transformed to have unit size.…”

Section: Face Recognition Reviewmentioning

confidence: 99%

Individual Discriminative Face Recognition Models Based on Subsets of Features

View full text Add to dashboard Cite

Abstract. The accuracy of data classification methods depends considerably on the data representation and on the selected features. In this work, the elastic net model selection is used to identify meaningful and important features in face recognition. Modelling the characteristics which distinguish one person from another using only subsets of features will both decrease the computational cost and increase the generalization capacity of the face recognition algorithm. Moreover, identifying which are the features that better discriminate between persons will also provide a deeper understanding of the face recognition problem. The elastic net model is able to select a subset of features with low computational effort compared to other state-of-the-art feature selection methods. Furthermore, the fact that the number of features usually is larger than the number of images in the data base makes feature selection techniques such as forward selection or lasso regression become inadequate. In the experimental section, the performance of the elastic net model is compared with geometrical and color based algorithms widely used in face recognition such as Procrustes nearest neighbor, Eigenfaces, or Fisherfaces. Results show that the elastic net is capable of selecting a set of discriminative features and hereby obtain higher classification rates.

show abstract

How effective are landmarks and their geometry for face recognition?

Cited by 121 publications

References 30 publications

Can facial metrology predict gender?

Can facial metrology predict gender?

Species differentiation of slipper orchids using color image analysis

Individual Discriminative Face Recognition Models Based on Subsets of Features

Contact Info

Product

Resources

About