Marco Morana

Human Activity Recognition Process Using 3-D Posture Data

In this paper, we present a method for recognizing human activities using information sensed by an RGB-D camera, namely the Microsoft Kinect. Our approach is based on the estimation of some relevant joints of the human body by means of the Kinect; three different machine learning techniques, i.e., K-means clustering, support vector machines, and hidden Markov models, are combined to detect the postures involved while performing an activity, to classify them, and to model each activity as a spatiotemporal evolution of known postures. Experiments were performed on Kinect Activity Recognition Dataset, a new dataset, and on CAD-60, a public dataset. Experimental results show that our solution outperforms four relevant works based on RGB-D image fusion, hierarchical Maximum Entropy Markov Model, Markov Random Fields, and Eigenjoints, respectively. The performance we achieved, i.e., precision/recall of 77.3% and 76.7%, and the ability to recognize the activities in real time show promise for applied use.

Motion sensors for activity recognition in an ambient-intelligence scenario

In recent years, Ambient Intelligence (AmI) has attracted a number of researchers due to the widespread diffusion of unobtrusive sensing devices. The availability of such a great amount of acquired data has driven the interest of the scientific community in producing novel methods for combining raw measurements in order to understand what is happening in the monitored scenario. Moreover, due the primary role of the end user, an additional requirement of any AmI system is to maintain a high level of pervasiveness. In this paper we propose a method for recognizing human activities by means of a time of flight (ToF) depth and RGB camera device, namely Microsoft Kinect. The proposed approach is based on the estimation of some relevant joints of the human body by using Kinect depth information. The most significative configurations of joints positions are combined by a clustering approach and classified by means of a multi-class Support Vector Machine. Then, Hidden Markov Models (HMMs) are applied to model each activity as a sequence of known postures. The proposed solution has been tested on a public dataset while considering four different configurations corresponding to some state-of-the-art approaches and results are very promising. Moreover, in order to maintain a high level of pervasiveness, we implemented a real prototype by connecting Kinect sensor to a miniature computer capable of real-time processing.

A framework for real-time Twitter data analysis

A framework for real-time Twitter data analysisWe propose improvements to the Soft Frequent Pattern Mining (SFPM) algorithmThe stream of tweets is organized in dynamic windows whose size depends both on the volume of tweets and timeThe set of keywords used to query Twitter is progressively refined to highlight the users point of viewComparisons with two state of the art systems Twitter is a popular social network which allows millions of users to share their opinions on what happens all over the world. In this work we present a system for real-time Twitter data analysis in order to follow popular events from the users perspective. The method we propose extends and improves the Soft Frequent Pattern Mining (SFPM) algorithm by overcoming its limitations in dealing with dynamic, real-time, detection scenarios. In particular, in order to obtain timely results, the stream of tweets is organized in dynamic windows whose size depends both on the volume of tweets and time. Since we aim to highlight the users point of view, the set of keywords used to query Twitter is progressively refined to include new relevant terms which reflect the emergence of new subtopics or new trends in the main topic. The real-time detection system has been evaluated during the 2014 FIFA World Cup and experimental results show the effectiveness of our solution.

A Data Association Algorithm for People Re-identification in Photo Sequences

In this paper, a new system is presented to support the user in the face annotation task. Every time a photo sequence becomes available, the system analyses it to detect and cluster faces in set corresponding to the same person. We propose to model the problem of people re-identification in photos as a data association problem. In this way, the system takes advantage from the assumption that each person can appear at most once in each photo. We propose a fully automated method for grouping facial images, the method does not require any initialization neither a priori knowledge of the number of persons that are in the photo sequence. We compare the results obtained with our method and with standard clustering methods on three personal collections and on a publicly available dataset.

Real-time detection of twitter social events from the user's perspective

Over the last 40 years, automatic solutions to analyze text documents collection have been one of the most attractive challenges in the field of information retrieval. More recently, the focus has moved towards dynamic, distributed environments, where documents are continuously created by the users of a virtual community, i.e., the social network. In the case of Twitter, such documents, called tweets, are usually related to events which involve many people in different parts of the world. In this work we present a system for real-time Twitter data analysis which allows to follow a generic event from the users point of view. The topic detection algorithm we propose is an improved version of the Soft Frequent Pattern Mining algorithm, designed to deal with dynamic environments. In particular, in order to obtain prompt results, the whole Twitter stream is split in dynamic windows whose size depends both on the volume of tweets and time. Moreover, the set of terms we use to query Twitter is progressively refined to include new relevant keywords which point out the emergence of new subtopics or new trends in the main topic. Tests have been performed to evaluate the performance of the framework and experimental results show the effectiveness of our solution.

Mobile Interface for Content-Based Image Management

People make more and more use of digital image acquisition devices to capture screenshots of their everyday life. The growing number of personal pictures raise the problem of their classification. Some of the authors proposed an automatic technique for personal photo album management dealing with multiple aspects (i. e., people, time and background) in a homogenous way. In this paper we discuss a solution that allows mobile users to remotely access such technique by means of their mobile phones, almost from everywhere, in a pervasive fashion. This allows users to classify pictures they store on their devices. The whole solution is presented, with particular regard to the user interface implemented on the mobile phone, along with some experimental results.

Enabling technologies on hybrid camera networks for behavioral analysis of unattended indoor environments and their surroundings

This paper presents a layered network architecture and the enabling technologies for accomplishing vision-based behavioral analysis of unattended environments. Specifically the vision network covers both the attended environment and its surroundings by means of hybrid cameras. The layer overlooking at the surroundings is laid outdoor and tracks people, monitoring entrance/exit points. It recovers the geometry of the site under surveillance and communicates people positions to a higher level layer. The layer monitoring the unattended environment undertakes similar goals, with the addition of maintaining a global mosaic of the observed scene for further understanding. Moreover, it merges information coming from sensors beyond the vision to deepen the understanding or increase the reliability of the system. The behavioral analysis is demanded to a third layer that merges the information received from the two other layers and infers knowledge about what happened, happens and will be likely happening in the environment. The paper also describes a case study that was implemented in the Engineering Campus of the University of Modena and Reggio Emilia, where our surveillance system has been deployed in a computer laboratory which was often unaccessible due to lack of attendance.

A data association approach to detect and organize people in personal photo collections

In this paper we present a method to automatically segment a photo sequence in groups containing the same persons. Many methods in literature accomplish to this task by adopting clustering techniques. We model the problem as the search for probable associations between faces detected in subsequent photos considering the mutual exclusivity constraint: a person can not be in a photo two times, nor two faces in the same photo can be assigned to the same group. Associations have been found considering face and clothing descriptions. In particular, a two level architecture has been adopted: at the first level, associations are computed within meaningful temporal windows (situations); at the second level, the resulting clusters are re-processed to find associations across situations. Experiments confirm our technique generally outperforms clustering methods. We present an analysis of the results on a public dataset, enabling future comparison, and on private collections.

Probabilistic Corner Detection for Facial Feature Extraction

After more than 35 years of resarch, face processing is considered nowadays as one of the most important application of image analysis. It can be considered as a collection of problems (i.e., face detection, normalization, recognition and so on) each of which can be treated separately. Some face detection and face recognition techniques have reached a certain level of maturity, however facial feature extraction still represents the bottleneck of the entire process. In this paper we present a novel facial feature extraction approach that could be used for normalizing Viola-Jones detected faces and let them be recognized by an appearance-based face recognition method. For each observed feature a prior distribution is computed and used as boost map to filter the Harris corner detector response producing more feature candidates on interest region while discarding external values. Tests have been performed on both AR and BioID database using approximately 1750 faces and experimental results are very encouraging.

Clustering techniques for personal photo album management

We propose a novel approach for the automatic representation of pictures achieving a more effective organization of personal photo albums. Images are analyzed and described in multiple representation spaces, namely, faces, background, and time of capture. Faces are automatically detected, rectified, and represented, projecting the face itself in a common low-dimensional eigenspace. Backgrounds are represented with low-level visual features based on an RGB histogram and Gabor filter bank. Faces, time, and background information of each image in the collection is automatically organized using a mean-shift clustering technique. Given the particular domain of personal photo libraries, where most of the pictures contain faces of a relatively small number of different individuals, clusters tend to be semantically significant besides containing visually similar data. We report experimental results based on a data set of about 1000 images where automatic detection and rectification of faces lead to approximately 400 faces. Significance of clustering has been evaluated, and results are very encouraging.