ENGLISH
La vitrine de diffusion des publications et contributions des chercheurs de l'ÉTS
RECHERCHER

Super-resolution pipeline for fast adjudication in watchlist screening

Téléchargements

Téléchargements par mois depuis la dernière année

Plus de statistiques...

Tayanov, Vitaliy, Granger, Éric, Bordallo, Miguel et Hadid, Abdenour. 2015. « Super-resolution pipeline for fast adjudication in watchlist screening ». In IEEE International Conference on Image Processing Theory, Tools and Applications (IPTA) (Orleans, France, Nov. 10-13, 2015) pp. 273-278. IEEE.
Compte des citations dans Scopus : 1.

[thumbnail of Granger E. 2015 11319 Super-resolution pipeline for fast adjudication in watchlist screening .pdf]
Prévisualisation
PDF
Granger E. 2015 11319 Super-resolution pipeline for fast adjudication in watchlist screening .pdf

Télécharger (1MB) | Prévisualisation

Résumé

Although still-to-video face recognition is an important function in watchlist screening, state-of-the-art systems often yield limited performance due to camera inter-operability and to variations in capture conditions. Therefore, the visual comparison of faces captured in unconstrained low-quality videos against a matching high-quality reference facial still image captured under controlled conditions is required in many surveillance applications to limit the number of costly false matches. To improve the visual appearance of faces captured in videos, this paper presents a new super-resolution (SR) pipeline that is suitable for fast adjudication of face-matches produced by an automated system. In this pipeline, face quality measures are used to rank and select face captures belonging to a facial trajectory, and multi-image SR iteratively enhances the appearance of a super-resolved face image. Face selection is optimized and registered using graphical models. Experiments with the Chokepoint dataset show that the proposed pipeline efficiently produces super-resolved face images by ranking best quality ROIs in a trajectory. To select the best face captures for SR, this pipeline exploits a strong correlation existing between pose and sharpness quality measurements.

Type de document: Compte rendu de conférence
Professeur:
Professeur
Granger, Éric
Affiliation: Génie de la production automatisée
Date de dépôt: 21 sept. 2015 17:48
Dernière modification: 08 janv. 2016 19:27
URI: https://espace2.etsmtl.ca/id/eprint/11319

Actions (Authentification requise)

Dernière vérification avant le dépôt Dernière vérification avant le dépôt