No abstract available.

Interactive face retrieval aims at finding target subjects in face databases through human and machine interaction, which involves user feedback based on human perception and machine similarity measure in feature spaces. In this article, we propose an attribute prototype learning method to tackle the semantic gap between human and machine in face perception for fast interactive face retrieval. We reformulate the theoretical explanation of the interactive retrieval model and develop the algorithm of the heuristic solution of the model. Each module of the prototype model is learned with a set of identity-related facial attributes. The outputs of the prototype modules form the semantic representation. To adapt the prototype models across different databases, we propose a transfer selection algorithm based on the coherence measurements in interactive face retrieval. Coherence analysis proves that the proposed attribute prototype representation can effectively narrow down the semantic gap even in the case of cross-database transfer learning. The prototype representation can effectively reduce the feature dimension in the retrieval process. Real user retrieval with the Bayesian relevance feedback model shows that attribute prototype space is superior to low-level feature space and proves that interactive retrieval with attribute prototype representation can converge fast in large face databases.

This paper presents several improvements of [email protected], an image sharing and retrieval application for identifying plants [6]: (i) ported to most android platforms (ii) three times more data (iii) exploiting metadata as well as visual content in the identification process (iv) a new multi-plant-organ, multi-image and multi-feature merging strategy with separate indexes for each visual feature (v) integrating cross-languages functions. This paper also presents the new results achieved by our system in the ImageCLEF 2013 plant identification task and in real-world user trials.

We study fine-grained categorization, the task of distinguishing among (sub)categories of the same generic object class (e.g., birds), focusing on determining botanical species (leaves and orchids) from scanned images. The strategy is to focus attention around several vantage points, which is the approach taken by botanists, but using features dedicated to the individual categories. Our implementation of the strategy is based on {\it vantage feature frames}, a novel object representation consisting of two components: a set of coordinate systems centered at the most discriminating local viewpoints for the generic object class and a set of category-dependent features computed in these frames. The features are pooled over frames to build the classifier. Categorization then proceeds from coarse-grained (finding the frames) to fine-grained (finding the category), and hence the vantage feature frames must be both detectable and discriminating. The proposed method outperforms state-of-the art algorithms, in particular those using more distributed representations, on standard databases of leaves.

There has recently been increasing interest in using advanced computer vision techniques for automatic plant identification. Most of the approaches proposed are based on an analysis of leaf characteristics. Nevertheless, two aspects have still not been well exploited: (1) domain-specific or botanical knowledge (2) the extraction of meaningful and relevant leaf parts. In this paper, we describe a new automated technique for leaf image retrieval that attempts to take these particularities into account. The proposed method is based on local representation of leaf parts. The part-based decomposition is defined and usually used by botanists. The global image query is a combination of part sub-images queries. Experiments carried out on real world leaf images, the Pl@ntLeaves scan images (3070 images totalling 70 species), show an increase in performance compared to global leaf representation.

Unravelling mysteries of the diversity of the plant community is a crucial issue both for the development of many botanical industries as well as for the conservation of ecosystem biodiversity. Traditionally, botanists have proposed detailed dichotomous key descriptions (called also characters or concepts) about the morphology of plants and particularly of leaves that allow them to construct relationships between different plants and between plants and their environment. However, extracting these concepts is complicated, painstaking and can only be carried out by experts. One way to accelerate and broaden the use of these concepts is to automatically extract them directly from images. In this paper, we focus on one of the most basic and important concepts: the leaf arrangement. According to this concept, leaves are divided into four categories: simple, pinnnately compound, palmately compound and compound trifoliate. To accomplish this task, we follow a hierarchical scheme, reducing ambiguity between categories from the most different shapes to the most similar ones. The choice of appropriate features is performed based on botanical observations and validated by a statistical study. The method was tested on real world leaf images (the Pl@ntLeaves scans). Experimental results show its robustness for a high number of leaf species (70 species) and even in the presence of some distortions (such as rotation and partial leaf overlapping).

Automatic retrieval tools are becoming increasingly important in botany and agriculture due to the growing interest in biodiversity and the ongoing shortage of skilled taxonomists. Our work is motivated by a botanical field scenario where the basic unit of observation is a plant. We describe a novel, image-based retrieval system for both educational and decision-making purposes. Given multiple leaf images of the same plant, the algorithm displays a ranked list of the most relevant species, along with a varied set of representative images from each estimated species. We focus on leaves but the strategy is generic, based on a hierarchical representation of latent variables called identification keys (IdKeys) which embody domain knowledge about taxonomy and landmarks. For each query image, keys are estimated sequentially, proceeding from landmarks to the genus and finally to an estimated set of species. The results over multiple queries are then collated into a single ranked list of species. Experiments demonstrate that the proposed approach achieves excellent performance on several databases of uncluttered leaf images as well as providing an instructive interface for measuring diversity and identifying new species.

[email protected] is an image sharing and retrieval application for the identification of plants, available on iPhone and iPad devices. Contrary to previous content-based identification applications it can work with several parts of the plant including flowers, leaves, fruits and bark. It also allows integrating user's observations in the database thanks to a collaborative workflow involving the members of a social network specialized on plants. Data collected so far makes it one of the largest mobile plant identification tool.

How-to train effective classifiers on huge amount of multimedia data is clearly a major challenge that is attracting more and more research works across several communities. Less efforts however are spent on the counterpart scalability issue: how to apply big trained models efficiently on huge non annotated media collections ? In this paper, we address the problem of speeding-up the prediction phase of linear Support Vector Machines via Locality Sensitive Hashing. We propose building efficient hash based classifiers that are applied in a first stage in order to approximate the exact results and filter the hypothesis space. Experiments performed with millions of one-against-one classifiers show that the proposed hash-based classifier can be more than two orders of magnitude faster than the exact classifier with minor losses in quality.

Automatic plant identification is a relatively new research area in computer vision that has increasingly attracted high interest as a promising solution for the development of many botanical industries and for the success of biodiversity conservation. Most of the approaches proposed are based on the analysis of morphological properties of leaves. They have applied several well-known generic shape descriptors. Nevertheless, faced with the large amount of leaf species, botanical knowledge, especially about leaf parts (petiole, blade and their insertion point) is important to enhance their precision, hence, a crucial need to extract them from image. In this paper, we propose a fully automatic approach for petiole detection, based on the concept of local translational symmetry, which is applied to a some regions of the leaf. These regions are chosen w.r.t their size (small) taking into account the large diversity of leaf morphology (compound, oblong, orbicular). This method has been tested on two datasets and has provided more than 90% of correct detections.

The problem of automatic leaf identification is particularly challenging because, in addition to constraints derived from image processing such as geometric deformations (rotation, scale, translation) and illumination variations, it involves difficulties arising from foliar properties. These include two main aspects: the first is the enormous number and diversity of leaf species and the second, which is relevant to some special species, is the high inter-species and the low intra-species similarity. In this paper, we present a novel boundary-based approach that attempts to overcome the most of these constraints. This method has been compared to results obtained in the image CLEF 2011 plant identification task. The main advantage of this first benchmark edition is that different image retrieval techniques were tested and a crowd-sourced leaf dataset was used. Our method provides the best classification rate for scan and scan-like pictures. Besides its high accuracy, our method satisfies real-time requirements with a low computational cost.

Efficiently constructing the K-Nearest Neighbor Graph (K-NNG) of large and high dimensional datasets is crucial for many applications with feature-rich objects, such as images or other multimedia content. In this paper we investigate the use of high dimensional hashing methods for efficiently approximating the K-NNG, notably in distributed environments. We first discuss the importance of balancing issues on the performance of such approaches and show why the baseline approach using Locality Sensitive Hashing does not perform well. Our new KNN-join method is based on RMMH, a recently introduced hash function family based on randomly trained classifiers. We show that the resulting hash tables are much more balanced and that the number of resulting collisions can be greatly reduced without degrading quality. We further improve the load balancing of our distributed approach by designing a parallelized local join algorithm, implemented within the MapReduce framework.

Understanding the results returned by automatic visual concept detectors is often a tricky task making users uncomfortable with these technologies. In this paper we attempt to build humanly interpretable visual models, allowing the user to visually understand the underlying semantic. We therefore propose a supervised multiple instance learning algorithm that selects as few as possible discriminant local features for a given object category. The method finds its roots in the lasso theory where a L1-regularization term is introduced in order to constraint the loss function, and subsequently produce sparser solutions. Efficient resolution of the lasso path is achieved through a boosting-like procedure inspired by BLasso algorithm. Quantitatively, our method achieves similar performance as current state-of-the-art, and qualitatively, it allows users to construct their own model from the original set of patches learned, thus allowing for more compound semantic queries.

Organizing media according to real-life events is attracting interest in the multimedia community. Event-centric indexing approaches are very promising for discovering more complex relationships between data. In this paper we introduce a new visual-based method for retrieving events in photo collections, typically in the context of User Generated Contents. Given a query event record, represented by a set of photos, our method aims to retrieve other records of the same event, typically generated by distinct users. Similarly to what is done in state-of-the-art object retrieval systems, we propose a two-stage strategy combining an efficient visual indexing model with a spatiotemporal verification re-ranking stage to improve query performance. For efficiency and scalability concerns, we implemented the proposed method according to the MapReduce programming model using Multi-Probe Locality Sensitive Hashing. Experiments were conducted on LastFM-Flickr dataset for distinct scenarios, including event retrieval, automatic annotation and tags suggestion. As one result, our method is able to suggest the correct event tag over 5 suggestions with a 72% success rate.

This demo presents a crowdsourcing web application dedicated to the access of botanical knowledge through automated identification of plant species by visual content. Inspired by citizen sciences, our aim is to speed up the collection and integration of raw botanical observation data, while providing to potential users an easy and efficient access to this botanical knowledge. The result presented during the demo is an enjoying application where anyone can play to shoot fresh cut leaves and observe the relevance of species suggested in spite of various visual difficult queries.

Query by visual example (QBVE) has been widely exploited in image retrieval. If starting image is missing, the query by visual thesaurus paradigm allows the user to compose his mental query image through visual patches summarizing the region database. Researches on the human visual system have provided considerable evidence that the color and texture must be processed separately. In this paper, we propose to enrich the paradigm of mental image search by constructing a hierarchical visual thesaurus of the regions provided by a new region labeling criterion into homogeneous and textured regions for boosting the object recognition. The new labeling criterion is based on the spatial dispersion of interest points in the region. Our point based criterion has been validated on a satellite image database. We can prove that our approach is able to retrieve complex concepts better than describing homogeneous and textured regions with the same visual feature.

In the context of computer-assisted plant identification we are facing challenging information retrieval problems because of the very high within-class variability and of the limited number of training examples. To address these problems, we suggest a new interactive learning approach that combines similarity-based retrieval and re-ranking by SVM using local feature distributions. This approach leads to improved sample selection, allowing to obtain better results.

The bag-of-visual-words is a popular representation for images that has proven to be quite effective for automatic annotation. In this paper, we extend this representation in order to include weak geometrical information by using visual word pairs. We show on a standard benchmark dataset that this new image representation improves significantly the performances of an automatic annotation system.

This paper presents an efficient local features boosting strategy for interactive objects retrieval tasks such as on-line supervised learning or relevance feedback. The prediction time complexity of most existing methods is indeed usually linear in dataset size since the retrieval works by applying a trained classifier on the images of the dataset one by one. In our method, the trained classifier can be computed directly on the whole dataset in sublinear time thanks to distance-based weak classifiers. The idea is to speed-up drastically the prediction of each weak classifier on the whole dataset by performing approximate range queries with an efficient similarity search structure. Experiments on Caltech 256 dataset show that the technique is up to 250 times faster than the naive exhaustive method. Thanks to this efficiency improvement, we developed a relevance feedback mechanism on image regions freely selected by the user and we show how it improves the effectiveness of the retrieval.

In this paper, we present a probabilistic framework for edge and region grouping using conditional random field. Our model is built on a hybrid adjacency graph of atomic region and contour primitives. Unary and pairwise potentials that capture similarity, proximity and curvilinear continuity are defined. Similarity, for both region and edge cues, is measured by likelihood ratios learned from a human labeled ground truth. We use a stochastic graph partition algorithm, Swendsen-Wang Cut, to perform inference on this model. Experimental results are shown on gray-scale natural images.

We propose a non-homogeneous conditional random field (CRF) built over an adjacency graph of superpixels for contextual region grouping. Our model includes spatially dependent potentials that capture contextual interactions of the data as well as the labels. Both superpixels and segments are described with local statistics which take into account their contexts in the image. This results the non-homogeneity of the fields which improves the region grouping process of natural images. In our energy formulation, the similarity is measured by a likelihood ratio learned from a human labeled ground truth. The inference is performed using a cluster sampling method, the Swendsen-Wang cut algorithm. Results are shown on various natural images.

Querying by visual thesaurus (VT) is a novel paradigm for content-based image retrieval approaches for it gives the user the possibility, in case of inappropriate starting example, to compose his query by arranging the visual patches of the starting "page zero" according to his mental image. A refinement of the willed results can be achieved by inducing a spatial description within the retrieval procedure. This paper presents a novel approach to model the spatial relations between the visual patches.We define the weighted angle spatial histogram (WASH) that combines the angular computation between pairs of regions of interest and their respective topological regularity/irregularity. WASH has shown great robustness to region shape and scale in the image because segmented regions are considered as a composition of elementary relevant and minor subregions. We tested our approach on generic database, and we compared it with other state-of-the-art techniques.

This paper presents applications for dealing with videos on the Web, using an efficient technique for video copy detection in large archives. Managing videos on the Web is the source of two exciting challenges: the respect of the copyright and the linkage of multiple videos. We present a technique called ViCopT for video copy tracking which is based on labels of behavior of local descriptors computed along video. The results obtained on large amount of data (270 hours of videos from the Internet) are very promising, even with a large video database (700 hours): ViCopT displays excellent robustness to various severe signal transformations, making it able to identify copies accurately from highly similar videos, as well as to link similar videos, in order to reduce redundancy or to gather the metadata associated. Finally, we also show that ViCopT goes further by detecting segments having the same background, with the aim of linking videos of the same category, like forecast weather programs or particular TV shows.

As the resolution of remote-sensing imagery increases, the full complexity of the scenes becomes increasingly difficult to approach. User-defined classes in large image databases are often composed of several groups of images and span very different scales in the space of low-level visual descriptors. The interactive retrieval of such image classes is then very difficult. To address this challenge, we evaluate here, in the context of satellite image retrieval, two general improvements for relevance feedback using support vector machines (SVMs). First, to optimize the transfer of information between the user and the system, we focus on the criterion employed by the system for selecting the images presented to the user at every feedback round. We put forward an active-learning selection criterion that minimizes redundancy between the candidate images shown to the user. Second, for image classes spanning very different scales in the low-level description space, we find that a high sensitivity of the SVM to the scale of the data brings about a low retrieval performance. We argue that the insensitivity to scale is desirable in this context, and we show how to obtain it by the use of specific kernel functions. Experimental evaluation of both ranking and classification performance on a ground-truth database of satellite images confirms the effectiveness of our approach

The visual thesaurus is a new query approach when no starting image is available. It is a concise representation of all similar regions in a panel of visual patches; the user arranges the visual patches according to his mental target image. The construction of the visual thesaurus needs a reliable region description and a clustering algorithm that reflects the variety of the database. In this paper, we develop a new region description schema based on Harris color points of interest. We also evaluate the relevance of several multi-dimensional matching metrics when measuring the similarity between regions described by variable signature dimensions. We outline the need of clustering to speed up the computation process as well. Moreover, we adopted the relational clustering algorithm to categorize regions according to Harris points of interest features. Generated clusters are represented by prototypes that compose the "page zero" of the visual thesaurus. We tested our approach on generic database, the relevance of obtained clusters is evaluated subjectively.

This paper presents an efficient approach for copies detection in a large videos archive consisting of several hundred of hours. The video content indexing method consists of extracting the dynamic behavior on the local description of interest points and further on the estimation of their trajectories along the video sequence. Analyzing the low-level description obtained allows to highlight trends of behaviors and then to assign a label of behavior to each local descriptor. Such an indexing approach has several interesting properties: it provides a rich, compact and generic description, while labels of behavior provide a high-level description of the video content. Here, we focus on video Content Based Copy Detection (CBCD). Copy detection is problematic as similarity search problem but with prominent differences. To be efficient, it requires a dedicated on-line retrieval method based on a specific voting function. This voting function must be robust to signal transformations and discriminating versus high similarities which are not copies. The method we propose in this paper is a dedicated on-line retrieval method based on a combination of the different dynamic contexts computed during the off-line indexing. A spatio-temporal registration based on the relevant combination of detected labels is then applied. This approach is evaluated using a huge video database of 300 hours with different video tests. The method is compared to a state-of-the art technique in the same conditions. We illustrate that taking labels into account in the specific voting process reduces false alarms significantly and drastically improves the precision.

Traditional clustering algorithms usually rely on a pre-defined similarity measure between unlabelled data to attempt to identify natural classes of items. When compared to what a human expert would provide on the same data, the results obtained may be disappointing if the similarity measure employed by the system is too different from the one a human would use. To obtain clusters fitting user expectations better, we can exploit, in addition to the unlabelled data, some limited form of supervision, such as constraints specifying whether two data items belong to a same cluster or not. The resulting approach is called semi-supervised clustering. In this paper, we put forward a new semi-supervised clustering algorithm, pairwise-constrained competitive agglomeration: clustering is performed by minimizing a competitive agglomeration cost function with a fuzzy term corresponding to the violation of constraints. We present comparisons performed on a simple benchmark and on an image database

We introduce in this paper a new formulation of the regularized fuzzy c-means (FCM) algorithm which allows us to find automatically the actual number of clusters. The approach is based on the minimization of an objective function which mixes, via a particular parameter, a classical FCM term and a new entropy regularizer. The main contribution of the method is the introduction of a new exponential form of the fuzzy memberships which ensures the consistency of their bounds and makes it possible to interpret the mixing parameter as the variance (or scale) of the clusters. This variance closely related to the number of clusters, provides us with an intuitive and an easy to set parameter. We will discuss the proposed approach from the regularization point-of-view and we will demonstrate its validity both analytically and experimentally. We will show an extension of the method to nonlinearly separable data. Finally, we will illustrate preliminary results both on simple toy examples as well as database categorization problems

Histogram intersection (HI) kernel has been recently introduced for image recognition tasks. The HI kernel is proved to be positive definite and thus can be used in support vector machine (SVM) based recognition. Experimentally, it also leads to good recognition performances. However, its derivation applies only for binary strings such as color histograms computed on equally sized images. In this paper, we propose a new kernel, which we named generalized histogram intersection (GHI) kernel, since it applies in a much larger variety of contexts. First, an original derivation of the positive definiteness of the GHI kernel is proposed in the general case. As a consequence, vectors of real values can be used, and the images no longer need to have the same size. Second, a hyper-parameter is added, compared to the HI kernel, which allows us to better tune the kernel model to particular databases. We present experiments which prove that the GHI kernel outperforms the simple HI kernel in a simple recognition task. Comparisons with other well-known kernels are also provided.

When using relevance feedback for the interactive categorization of images, the strategy employed by the system to select images to be presented to the user is of paramount importance for overall performance. Using SVM-based relevance feedback, we present a new selection criterion, based on the active learning principle, that minimizes redundancy between the candidate images shown to the user at every round. We also emphasize the fact that insensitivity to the scale of the target classes in the description space is an important quality of the learner in the interactive categorization context and we propose specific kernel functions to achieve this. Experimental results on several image databases confirm the attractiveness of our suggestions.

As image collections become ever larger, effective access to their content requires a meaningful categorization of the images. Such a categorization can rely on clustering methods working on image features, but should greatly benefit from any form of supervision the user can provide, related to the visual content. Semi-supervised clustering - learning from both labelled and unlabelled data - has consequently become a topic of significant interest. In this paper we present a new semi-supervised clustering algorithm, pairwise-constrained competitive agglomeration, which is based on a fuzzy cost function that takes pairwise constraints into account.

Content-based image retrieval has shown to be more and more useful for several application domains, from audiovisual media to security. As content-based retrieval became mature, different scientific applications were revealed client for such methods. More recently, botanical applications generated very large image collections and then became very demanding content-based visual similarity computation. In this paper, we describe low-level feature extraction for visual appearance comparison between genetically modified plants for gene expression studies.

Kernel based methods such as support vector machine (SVM) has provided successful tools for solving many recognition problems. One of the reasons of this success is the use of kernels. Positive definiteness has to be checked for kernels to be suitable for most of these methods. For instance for SVM, the use of a positive definite kernel insures that the optimized problem is convex and thus the obtained solution is unique. Alternative class of kernels called conditionally positive definite have been studied for a long time from the theoretical point of view and have drawn attention from the community only in the last decade. We propose a new kernel, named log kernel, which seems particularly interesting for images. Moreover, we prove that this new kernel is a conditionally positive definite kernel as well as the power kernel. Finally, we show from experimentations that using conditionally positive definite kernels allows us to outperform classical positive definite kernels

We introduce the intermediate matching (IM) kernel for SVM-hased object recognition. The IM kernel operates on a feature space of vector sets where each image is represented by a set of local features. Matching algorithms have proved to be efficient for such types of features. Nevertheless, kernelizing the matching for SVM does not lead to positive definite kernels. The IM kernel overcomes this drawback, as it mimics matching algorithms while being positive definite. The IM kernel introduces an intermediary set of so-called virtual local features. These select the pairs of local features to be matched. Comparisons with the matching kernel shows that the IM kernels leads to similar performances.

Clustering is considered as one of the most important tools to organize and analyze large multimedia databases. Most existing clustering techniques assume that the clusters have well-defined shapes (spherical or ellipsoidal). Thus, they are not suitable for image database categorization where images are usually mapped to high-dimensional feature vectors, and it is hard to even guess the shape of the clusters in the feature space. In this paper, we assume that the high dimensional object signature can be modeled by a fuzzy set and we introduce an algorithm to cluster these sets. First, we define a measure to assess the dissimilarity between two fuzzy sets. Then, we integrate this measure into our synchronization-based clustering approach. The resulting algorithm, called SyMP/sub FD/ is robust to noise and outliers, determines the number of clusters in an unsupervised manner, and identifies clusters of arbitrary shapes. The robustness of SyMP/sub FD/ is an intrinsic property of the synchronization mechanism. To identify clusters of various shapes, SyMP/sub FD/ models each cluster by an ensemble of fuzzy sets. Clusters with simple shapes would be modeled by few sets while clusters with more complex shapes would require a larger number of sets. The performance of the proposed algorithm is illustrated by using it to categorize a collection of images, where each image is described by a fuzzy set representing its color distribution.

To allow efficient browsing of large image collection, we have to provide a summary of its visual content. We present in this paper a new robust approach to categorize image databases: Adaptive Robust Competition with Proximity-Based Merging (ARC-M). This algorithm relies on a non-supervised database categorization, coupled with a selection of prototypes in each resulting category. Each image is represented by a high-dimensional vector in the feature space. A principal component analysis is performed for every feature to reduce dimensionality. Then, clustering is performed in challenging conditions by minimizing a Competitive Agglomeration objective function with an extra noise cluster to collect outliers. Agglomeration is improved by a merging process based on cluster proximity verification.

We present a novel framework for intelligent search and retrieval by image content composition. Very different from the existing query-by-example paradigm, logical queries are expressed using categories of similar regions without any starting example region. The set of region category representatives constitutes the "photometric region thesaurus" of the image database. Logical composition of region categories expresses the presence and absence of certain types of regions in images to retrieve and allows to integrate visual semantics in the search. Resulting indexing and retrieval implementation turns out to be simple and very fast even on very complex query compositions and large image database. It was tested on a database of 9,995 images from the Corel Photostock.

The two major problems raised by a region-based image retrieval system are the automatic definition and description of regions. We first present a technique of unsupervised coarse detection of regions which improves their visual specificity. The segmentation scheme is based on the classification of local distributions of quantized colors (LDQC). The competitive agglomeration (CA) classification algorithm is used which has the advantage to automatically determine the optimal number of classes. Then, considering that region description which must be finer for regions than for images, we propose a region descriptor of fine color variability: the adaptive distribution of color shades (ADCS). Compared to existing color descriptors, the high color resolution of ADCS improves the perceptual similarity of retrieved regions.

For content-based image retrieval (CBIR), traditional approaches of image matching involve global descriptions of the color image. When considering particular tasks like object recognition or partial queries, more local characterizations must be employed. In this context, image description based on points of interest appear best adapted. The point characterization which proved reliable is based on combinations of the Hilbert's differential invariants. For gray value images, such a description used to be considered up to third order. Generalizations to color images were previously proposed for stereovision and image retrieval. Some of them propose to consider the invariants only at first order, while others consider higher order invariants and compute some combinations of them to achieve illumination changes invariance. We discuss the advantages and drawbacks of these different choices, with the aim of proposing an optimal use of color points of interest for image retrieval.

We describe a new face detection algorithm based on a hierarchy of support vector classifiers (SVM) designed for efficient computation. The hierarchy serves as a platform for a coarse-to-fine search for faces: most of the image is quickly rejected as "background" and the processing naturally concentrates on regions containing faces and face-like structures. The hierarchy is tree-structured: In proceeding from the root to the leaves, the SVM gradually increase in complexity (measured by the number of support vectors) and discrimination (measured by the false alarm rate), but decrease in the level of invariance. Reduced complexity is achieved by clustering support vectors and shifting the decision boundary in order to satisfy a "conservation hypothesis" that preserves positive responses from the original set of support vectors. The computation is organized as a depth-first search and cancel strategy. The gain in efficiency is enormous.

Self organizing maps are efficient for dimension reduction and data clustering. We propose the use of the Kohonen topologic map for fingerprint pattern classification. The learning process takes into account the large intra-class diversity and the continuum of fingerprint pattern types. After a brief introduction to fingerprint domain-specific knowledge and the expert approach, we present an original and intuitive description of the algorithm. For a classification based on the global shape of the fingerprint, we adopted a suitable feature space. Indeed we obtained 88% correct classification on a database composed of 1600 NIST fingerprints.

The utility of face recognition for multimedia indexing is enhanced by using accurate detection and alignment of salient invariant face features. The face recognition can be performed using template matching or a feature-based approach, but both these methods suffer from occlusion and require an a priori model for extracting information. To avoid these drawbacks, we present a complete scheme for face recognition based on salient feature extraction in challenging conditions, which is performed without an a priori or learned model. These features are used in a matching process that overcomes occlusion effects using the dynamic space warping which aligns each feature in the query image, if possible, with its corresponding feature in the gallery set. Thus, we make face recognition robust to low frequency variations (like the presence of occlusion, etc) as well as to high frequency variations (like expression, gender, etc). A maximum likelihood scheme is used to make the recognition process more precise, as is shown in the experiments.

This contribution proposes a fuzzy approach to color image filtering by the fuzzy modeling of the concept of color credibility. Based on the perceptual notion of color resemblance, the colors are modeled as fuzzy sets in the CIELAB color space. The filtering principle is to select at the filters output the color that is the most credible with respect to the rest of the colors within the filtering window. Although the approach does not make any assumption on the desired filter type, the result is similar to a vector median-type filter.

We focus on the problem of unsupervised clustering which allows automatic setting of optimal clusters number. We present a generalization of the competitive agglomeration clustering algorithm first introduced by Frigui et al. (1997). This generalization is inspired by the regularization theory and suggests a new schema for using various cluster validity criteria proposed in the literature. As a consequence of this generalization, we introduce new objective clustering functions, and present their associated optimal solutions. We present an application of this competitive clustering schema to color image segmentation in order to perform partial queries in the context of image retrieval by content. In this case, each pixel is represented by the color distribution in its vicinity. The clustering algorithm has to incorporate an appropriate distance measure to compare feature vectors similarity.

This paper proposes a novel approach to color texture characterization and classification. Rather than developing new textural features, we propose to derive a family of new, reduced dimensionality color spaces named P/sub 1/P/sub 2/, that allow a good classification performance by the use of classical energy-distribution features, defined in a scalar spectral domain. The dimensionality reduction approach can be traced back to color constancy normalization and the reduced ordering principle and exhibits a strong perceptual background. We develop an adaption procedure for the selection of the proper color space within the new P/sub 1/P/sub 2/ family. The overall classification performance is very promising and the proposed methodology surmounts the current color texture characterization by features extracted from the luminance spectrum only.

A2thozgh pouleTful image TepT~entatiom have been proposed for cozienf-based image Tetieval, most of the cument systems aTe "tigid", i.e. they rettieve a fied set of images as Tesponse to a given que~and an image featuTe.We introduce S=fitaage, a useT-fiendly, genetic and flw-ble content-based image TetievaI system.Stiiatage wes the que~-by-sample appToach for ~ettieving images and integrate advanced ~eatuTessuch as image signatuTe combination, classification, multiple queti~and queq Refinement.The classic and advanced featuTes of SMiraage aTe detaild in the papeT.Stiiraage has been eztensive[y t~ted oz dozens of databas~and p~oauced ezcellent Tetm-evalTe-S-dts; a sample of Tetm"evaTau~ts h pr~ented here.

Recognition of handwritten digits is useful for several applications such as automatic bank cheques interpretation. Many problems occur, making this task quite difficult: digits may overlap, the removal of a base line may damage the digits, and noise quantization pixels may alter the digits shape and meaning, etc. Uncertainty modeling becomes essential to our work. This paper shows how robust fuzzy clustering techniques are suitable and useful for soft feature extraction and representation of a cheque's numerical value.