Facial soft biometric features for forensic face recognition

Abstract

El acceso a la versión del editor puede requerir la suscripción del recurso Access to the published version may require subscription AbstractThis paper proposes a functional feature-based approach useful for real forensic caseworks, based on the shape, orientation and size of facial traits, which can be considered as a soft biometric approach. The motivation of this work is to provide a set of facial features, which can be understood by non-experts such as judges and support the work of forensic examiners who, in practice, carry out a thorough manual comparison of face images paying special attention to the similarities and differences in shape and size of various facial traits. This new approach constitutes a tool that automatically converts a set of facial landmarks to a set of features (shape and size) corresponding to facial regions of forensic value. These features are furthermore evaluated in a population to generate statistics to support forensic examiners. The proposed features can also be used as additional information that can improve the performance of traditional face recognition systems. These features follow the forensic methodology and are obtained in a continuous and discrete manner from raw images. A statistical analysis is also carried out to study the stability, discrimination power and correlation of the proposed facial features on two realistic databases: MORPH and ATVS Forensic DB. Finally, the performance of both continuous and discrete features is analysed using different similarity measures. Experimental results show high discrimination power and good recognition performance, especially for continuous features. A final fusion of the best systems configurations achieves rank 10 match results of 100% for ATVS database and 75% for MORPH database demonstrating the benefits of using this information in practice.

Method to assess the temporal persistence of potential biometric features: Application to oculomotor, gait, face and brain structure databases

“…The same can be said of a t-value from a linear regression, i.e., this is also invariant to linear transformations. Another study created their own “stability index” [11], which is based on the average absolute deviation from either the mode or the mean (for discrete versus continuous measures, respectively). Why these researchers chose to invent their own stability index, rather than use the already well established Kendall Coefficient of Concordance (first described in 1939 [12], for continuous, not normally distributed data) or the ICC (first described in 1925, [4], for continuous, normally distributed data) is not clear.…”

Semantic Face Signatures: Recognizing and Retrieving Faces by Verbal Descriptions

“…9). The approach proposed in [16] used shape and size features of facial regions as soft biometrics and achieved rank-10 accuracies of 100% and 75% with the relatively constrained ATVS and MORPH databases respectively, while the best EER achieved in [16] was 3.06% and 12.27% for ATVS and MORPH databases correspondingly. In [18], categorical facial attributes were proposed for active authentication on mobile devices and achieved an EER of 14% with the MOBIO dataset, which is composed of video footage acquired by mobile cameras for 152 subjects.…”

“…In [16], Tome traditional face recognition systems to improve recognition performance. Their experiments, which used the ATVS [16] and MORPH [17] databases, demonstrated the significant discriminative power of categorical soft biometrics and the performance gain that can result from using them in fusion with traditional face recognition systems. Samangouei et al [18] proposed a method for face authentication on mobile devices that is based on categorical (binary) attributes.…”

Aceptación del Reconocimiento Facial Como Medida de Vigilancia y Seguridad: Un Estudio Empírico en Chile

“…), y luego algunas medidas que describan esas regiones, en cambio, la categoría holística intenta identificar los rostros usando representaciones globales, es decir, descripciones basadas en la imagen completa, en vez de las características locales de la cara. Además, existen métodos híbridos que detectan puntos de referencia y luego aplican técnicas utilizadas por la categoría holística (Tome et al, 2015). En los últimos años, muchos países que han adoptado esta tecnología como medida de seguridad, se utilizan en aeropuertos o en áreas públicas y especialmente en lugares con una mayor posibilidad de eventos importantes según el criterio del vigilante.…”