Raffaella Lanzarotti

Precise eye localization through a general-to-specific model definition

We present a method for precise eye localization that uses two Support Vector Machines trained on properly selected Haar wavelet coefficients. The evaluation of our technique on many standard databases exhibits very good performance. Furthermore, we study the strong correlation between the eye localization error and the face recognition rate.

A face recognition system based on automatically determined facial fiducial points

In this paper, a completely automatic face recognition system is presented. The method works on color images: after having localized the face and the facial features, it determines 24 facial fiducial points, and characterizes them applying a bank of Gabor filters which extract the peculiar texture around them (jets). Recognition is realized measuring the similarity between the different jets. The system is inspired by the elastic bunch graph method, while it does no assumption on the scale, pose, and the background. Comparison with standard algorithms is presented and discussed.

PRECISE EYE AND MOUTH LOCALIZATION

The literature on the topic has shown a strong correlation between the degree of precision of face localization and the face recognition performance. Hence, there is a need for precise facial featu...

Automatic Facial Feature Extraction for Face Recognition

Facial feature extraction consists in localizing the most characteristic face components (eyes, nose, mouth, etc.) within images that depict human faces. This step is essential for the initialization of many face processing techniques like face tracking, facial expression recognition or face recognition. Among these, face recognition is a lively research area where it has been made a great effort in the last years to design and compare different techniques. In this chapter we intend to present an automatic method for facial feature extraction that we use for the initialization of our face recognition technique. In our notion, to extract the facial components equals to locate certain characteristic points, e.g. the center and the corners of the eyes, the nose tip, etc. Particular emphasis will be given to the localization of the most representative facial features, namely the eyes, and the locations of the other features will be derived from them. An important aspect of any localization algorithm is its precision. The face recognition techniques (FRTs) presented in literature only occasionally face the issue and rarely state the assumptions they make on their initialization; many simply skip the feature extraction step, and assume perfect localization by relying upon manual annotations of the facial feature positions. However, it has been demonstrated that face recognition heavily suffers from an imprecise localization of the face components. This is the reason why it is fundamental to achieve an automatic, robust and precise extraction of the desired features prior to any further processing. In this respect, we investigate the behavior of two FRTs when initialized on the real output of the extraction method.

Fiducial point localization in color images of face foregrounds

Abstract We describe a method for the automatic identification of facial features (eyes, nose, mouth and chin) and the precise localization of their fiducial points (e.g. nose tip, mouth and eye corners) in color images of face foregrounds. The algorithm requires as input 2D color images, representing face foregrounds with homogeneous background; it is scale-independent, it deals with either frontal, rotated (up to 30°) or slightly tilted (up to 10°) faces, and it is robust to different facial expressions, requiring the mouth closed and the eyes opened, and no wearing glasses. The method proceeds with subsequent refinements: first, it identifies the sub images containing each feature, afterwards, it processes the single features separately by a blend of techniques which use both color and shape information. The system has been tested on three databases: the XM2VTS database, the University of Stirling database, and ours, for a total of 1650 images. The obtained results are described quantitatively and discussed.

A feature-based face recognition system

A completely automatic face recognition system is presented. The method works on color and gray level images: after having localized the face and the facial features, it determines 16 facial fiducial points, and characterizes them by applying a bank of filters which extract the peculiar texture around them (jets). Recognition is realized by measuring the similarity between the different jets. The system is inspired by the elastic bunch graph method, but the fiducial point localization does not require any manual setting or operator intervention.

Face and facial feature localization

In this paper we present a general technique for face and facial feature localization in 2D color images with arbitrary background. In a previous work we studied an eye localization module, while here we focus on mouth localization. Given in input an image that depicts a sole person, first we exploit the color information to limit the search area to candidate mouth regions, then we determine the exact mouth position by means of a SVM trained for the purpose. This component-based approach achieves the localization of both the faces and the corresponding facial features, being robust to partial occlusions, pose, scale and illumination variations. We report the results of the separate modules of the single feature classifiers and their combination on images of several public databases.

A face recognition system based on local feature analysis

In this paper a completely automatic face recognition system is presented. The system is inspired by the elastic bunch graph method, but the fiducial point localization is completely different and does not require any operator intervention. Each fiducial point is characterized applying a bank of filters which extracts the peculiar texture around it (jet). The performances of the steerable Gaussian first derivatives basis filters are compared to the ones of the Gabor wavelet transform, showing similar results when images of faces in approximately the same pose are compared.

ECG compression retaining the best natural basis k-coefficients via sparse decomposition

Abstract A novel and efficient signal compression algorithm aimed at finding the sparsest representation of electrocardiogram (ECG) signals is presented and analyzed. The idea behind the method relies on basis elements drawn from the initial transitory of a signal itself, and the sparsity promotion process applied to its subsequent blocks grabbed by a sliding window. The saved coefficients rescaled in a convenient range, quantized and compressed by a lossless entropy-based algorithm. Experiments on signals extracted from the MIT-BIH Arrhythmia database show that the method achieves in most of the cases very high performance.

Automatic features detection for overlapping face images on their 3D range models

We describe an algorithm for automatic features detection in 2D color images of human faces. The algorithm proceeds with subsequent refinements. First, it identifies the sub-images containing each feature (eyes, nose and lips). Afterwards, it processes the single features separately by a blend of techniques which use both color and shape information. The method does not require any manual setting or operator intervention.