Publications

In this paper we present a people posture classificationapproach especially devoted to cope with occlusions. Inparticular, the approach aims at assessing temporal coherenceof visual data over probabilistic models. A mixed predictiveand probabilistic tracking is proposed: a probabilistictracking maintains along time the actual appearance ofdetected people and evaluates the occlusion probability; anadditional tracking with Kalman prediction improves the estimationof the people position inside the room. ProbabilisticProjection Maps (PPMs) created with a learning phaseare matched against the appearance mask of the track. Finally,an Hidden Markov Model formulation of the posturecorrects the frame-by-frame classification uncertainties andmakes the system reliable even in presence of occlusions.Results obtained over real indoor sequences are discussed.

In this paper we propose a system for automatic detection of shots in sport videos. Our work covers two main aspects: the first is robust shot detection in presence of fast object motion and camera operations. To this aim we propose a new algorithm, unique for both cuts and linear transitions detection, which only needs the tuning of two parameters. An extended comparison with four transition detection algorithms, representing the state of the art in literature, is reported. Examples with formula 1, basket, soccer and cycling videos are analyzed. The second aspect is an in depth discussion on the annotation of shots and transitions with the MPEG-7 standard.

Background/purpose: Atypical nevi (AN) share some dermoscopic features with early melanoma (MM), and computer elaboration of digital images could represent a useful support to diagnosis to assess automatically colors in AN, and to compare the data with those referring to clearly benign nevi (BN) and MMs. Methods: An image analysis program enabling the numerical description of color areas in melanocytic lesions was used on 459 videomicroscopic images, referring to 76 AN, 288 clearly BN and 95 MMs. Results: Black, white and blue-gray were more frequently found in AN than in clearly BN, but less frequently than in MMs. Color area values significantly differed between the three groups. Conclusion: The clinical-morphological interpretation of the numerical data, based on the mathematical description of the aspect and distribution of different color areas in different lesion types may contribute to the characterization of AN and their distinction from MMs.

Video annotation is typically performed by classifying video elements according to some pre-defined ontology of the video content domain. Ontologies are defined by establishing relationships between linguistic terms, that specify domain concepts at different abstraction levels. However, although linguistic terms are appropriate to distinguish event and object categories, they are inadequate when they must describe specific patterns of events or video entities. Instead, in these cases, pattern specifications are better expressed through visual prototypes that capture the essence of the event or entity. Pictorially enriched ontologies, that include visual concepts together with linguistic keywords, are therefore needed tosupport video annotation up to the level of detail of pattern specification. This paper presents pictorially enriched ontologies and provide a solution for their implementation in the soccer video domain. The pictorially enriched ontology is used both to directly assign multimedia objects to concepts, providing a more meaningful definition than the linguistics terms, and to extend the initial knowledge of the domain, adding subclasses of highlights or new highlight classes that were not defined in the linguistic ontology. Automatic annotation of soccer clips up to the pattern specification level using a pictorially enriched ontology is discussed.

Metadata extraction is one of the first tasks to be performed for automatic Digital Library annotation, and in particular shot detection has been widely explored in literature. While a lot of methods have been proposed for the detection of abrupt cuts, only a small number of them has explicitly addressed the problem of gradual transitions. In this paper we propose an algorithm that exploits a precise model of linear transition. Experimental results on Formula 1 car races videos show the robustness of this method. These test videos are characterized by extreme situations such as fast camera and objects motion and very different kinds of shots. The algorithm is able to estimate the exact length of the transition and an error score is also given as a fitness measure to the linear model, to discriminate true transitions from false detections. The final shot segmentation is delivered as an MPEG7 compliant output.

Video surveillance has a direct application in intelligent home automation or domotics (from the Latin word domus, that means “home”, and informatics). In particular, in-house video surveillance can provide good support for people with some difficulties (e.g. elderly or disabled people) living alone and with limited autonomy. A key aspect in video surveillance systems for domotics is that of analyzing behaviours of the monitored people. To accomplish this task, people must be detected and tracked, and their posture must be analyzed in order to model behaviours recognizing abrupt changes in it. Problems related to reliable software solutions are not completely solved, in particular luminance changes, shadows and frequent posture changes must be taken into account. Long-lasting occlusions are common due to the proximity of the cameras and the presence of furniture and doors that can often hide parts of a person’s body. For these reasons, a probabilistic and appearance-based tracking, particularly conceivable for people tracking and posture classification, has been developed. However, despite its effectiveness for long-lasting and large occlusions, this approach tends to fail whenever the person is monitored with multiple cameras and he appears in one of them already occluded. Different views provided by multiple cameras can be exploited to solve occlusions by warping known object appearance into the occluded view. To this aim, this paper describes an approach to posture classification based on projection histograms, reinforced by HMM for assuring temporal coherence of the posture.

This paper addresses auto-iris compensation. Auto-iris can be really troublesome for motion detection and tracking techniques based on background or frame differencing,since it can change quickly the average intensity of thecurrent frame. To cope with this, we introduced a two-step autoiris compensation approach in our traffic monitoring system. First, the auto-iris detection is based on the computation of the average of the luminance difference obtained by background suppression. Then, if an auto-iris is detected, the compensation phase is started. In this phase, the auto-iris’ behaviour is empirically modelled and, thus, compensated. Experimental results demonstrate the accuracy of the proposed approach, with both quantitative measures and visual analysis.

In vivo reflectance-mode confocal laser microscopy enables the visualization of the skin at quasi- histopathologic resolution. The aim of our study was to describe confocal features in melanocytic lesions, to evaluate their diagnostic significance for melanoma identification, to develop a simple algorithm useful for diagnostic purposes. A total of 102 consecutive melanocytic lesions (37 melanomas, 49 acquired nevi and 16 Spitz nevi), corresponding to lesions with equivocal aspects at clinical and dermoscopic inspection and excised in order to rule out a melanoma, were investigated by means of confocal microscopy (Vivascope 1000). In superficial layers the general pattern and the presence and aspects of pagetoid cells were evaluated. At basal cell layer dermal papilla features and cytological aspects, suggesting the presence of cellular atypia, were described. In dermal papilla, the presence and morphology of melanocytic nests and the presence and aspect of solitary cells were evaluated. Some features were more frequently observed in melanomas. In multivariate analysis 6 features appeared independently correlated with melanoma diagnosis. The presence of non edged dermal papillae, atypical cells in basal layers and isolated nucleated cells within dermal papilla were strongly correlated with melanoma diagnosis and were considered as major criteria (scored 2 points), whereas the presence of pagetoid cells, a widespread pagetoid infiltration in superficial layers and cerebriform nests in upper dermis were considered ‘‘minor’’ criteria (scored 1 point). A total score, ranging between 0 to 9, was obtained for each lesion and a ROC curve with an area under the curve of 0.951 was obtained on our dataset. In conclusion, characterization of confocal microscopy features of melanomas and nevi seems to improve diagnostic accuracy for difficult to diagnose melanocytic lesions.

In this paper, we present joint research activities in computer vision and sensor networks for a distributedsurveillance of urban parks. Distributed visual surveillance of urban environments is one of the most interesting scenarios in Ambient Intelligence; in addition, the automated monitoring of public parks, often crowded by children and aduits, is still a very difficult task due to the number of objects of interests. In this context, integrating the power of low cost sensors with the information provided by cameras can lead to a more reliable solution to people tracking in wide areas. Specifically, the deficiencies of one approach can be (at least partially) covered by the advantages of the other. The goal is to perform people tracking in parks (toachieve trackable parks - T-Parks), both in zones covered by overlapped cameras and afso, thanks to sensors, in areas not covered by any camera. In this paper, we propose a new technique for multi-camera people tracking based on a learning phase to automatically calibrate pairs of cameras and to build Areas of Field Views (AoFoVs) in order to establish consistent labelling of people. In addition, sensornetworks distributed at the borders of the AoFoV give an estimation of the probability of people overlapping, triggering specific algorithms of face detection or headcounting to identify the single person. The research ofT-Parks is part of a two-year Italian project called LAICA, intended to provide advanced services for citizens and public officers based on ambient intelligence technologies.