The BANCA Database and Evaluation Protocol

Abstract

Abstract. In this paper we describe the acquisition and content of a new large, realistic and challenging multi-modal database intended for training and testing multi-modal verification systems. The BANCA database was captured in four European languages in two modalities (face and voice). For recording, both high and low quality microphones and cameras were used. The subjects were recorded in three different scenarios, controlled, degraded and adverse over a period of three months. In total 208 people were captured, half men and half women. In this paper we also describe a protocol for evaluating verification algorithms on the database. The database will be made available to the research community through http://www.ee.surrey.ac.uk/Research/VSSP/banca.

On the Use of SIFT Features for Face Authentication

“…Some examples of the application of the SIFT algorithm to face images (database BANCA [1]) are shown in Fig. 3 and 4.…”

“…In this paper a first attempt to apply the SIFT for face classification is reported. The basic SIFT scheme is tested on a standard face database [1] with three different matching techniques.…”

Haar Local Binary Pattern Feature for Fast Illumination Invariant Face Detection

“…Fleuret 580 Non-uniform Images from real life situations were collected from the web, [3] illumination showing large variations in illumination, background and face size and slight variations in pose. For training, we used two internally created databases consisting of face and non-face images extracted from BANCA(Spanish Corpus) [1], Essex, Feret [13], ORL [15], Stirling and Yale [2] databases. For testing, we used 1) the standard XM2VTS database [11], [8], taking into account two cases, the Normal set with normal lighting conditions and the Darkened set with adverse or side illumination, 3) the BioID database [6] and 4) an additional database from Fleuret et al [3].…”

“…The second is the system based on MCT [4] which is one of the best performing systems representing the state of the art today. 1 The rest of the paper is organized as follows: we first introduce the proposed HLBP features in Sec. 2.…”

Face Description for Perceptual User Interfaces

“…A well known corpus used to evaluate recognition techniques is the FERET database [22] and more recently the Face Recognition Vendor Test. Veri cation approaches have their own framework, the BANCA protocol [2]. However, other facial descriptors which are particularly useful to describe unknown individuals, or to realize changes in human appearance during social interaction, have not excited the interest of the researchers similarly.…”