2011
Conference article
Open Access
LUCIA: An open source 3D expressive avatar for multimodal h.m.i.
Cosi P., Leone G. R., Paci G.
LUCIA is an MPEG-4 facial animation system developed at ISTC-CNR . It works on standard Facial Animation Parameters and speaks with the Italian version of FESTIVAL TTS. To achieve an emotive/expressive talking head LUCIA was build from real human data physically extracted by ELITE optotracking movement analyzer. LUCIA can copy a real human by reproducing the movements of passive markers positioned on his face and recorded by the ELITE device or can be driven by an emotional XML tagged input text, thus realizing a true audio/visual emotive/expressive synthesis. Synchronization between visual and audio data is very important in order to create the correct WAV and FAP files needed for the animation. LUCIA's voice is based on the ISTC Italian version of FESTIVAL-MBROLA packages, modified by means of an appropriate APML/VSML tagged language. LUCIA is available in two dif-ferent versions: an open source framework and the "work in progress" WebGL.
Cosi P., Leone G. R., Paci G.
LUCIA is an MPEG-4 facial animation system developed at ISTC-CNR . It works on standard Facial Animation Parameters and speaks with the Italian version of FESTIVAL TTS. To achieve an emotive/expressive talking head LUCIA was build from real human data physically extracted by ELITE optotracking movement analyzer. LUCIA can copy a real human by reproducing the movements of passive markers positioned on his face and recorded by the ELITE device or can be driven by an emotional XML tagged input text, thus realizing a true audio/visual emotive/expressive synthesis. Synchronization between visual and audio data is very important in order to create the correct WAV and FAP files needed for the animation. LUCIA's voice is based on the ISTC Italian version of FESTIVAL-MBROLA packages, modified by means of an appropriate APML/VSML tagged language. LUCIA is available in two dif-ferent versions: an open source framework and the "work in progress" WebGL.
2012
Conference article
Open Access
A webGL talking head for mobile devices
Benin A., Cosi P., Leone G. R.
Luciaweb is a 3D Italian talking avatar based on the new WebGL technology. WebGL is the standard programming library to develop 3D computer graphics inside the web browsers. In the last year we developed a facial animation system based on this library to interact with the user in a bimodal way. The overall system is a client-server application using the http protocol: we have a client (a browser or an app) and a web server. No software download and no plugin are required. All the software reside on the server and the visualization player is delivered inside the html pages that the client ask at the beginning of the connection. On the server side a software called AudioVideo Engine generates the phonemes and visemes information needed for the animation. The demo called Emotional Parrot shows the ability to reproduce the same input in different emotional states. This is the first WebGL software running on iOS device ever.
Benin A., Cosi P., Leone G. R.
Luciaweb is a 3D Italian talking avatar based on the new WebGL technology. WebGL is the standard programming library to develop 3D computer graphics inside the web browsers. In the last year we developed a facial animation system based on this library to interact with the user in a bimodal way. The overall system is a client-server application using the http protocol: we have a client (a browser or an app) and a web server. No software download and no plugin are required. All the software reside on the server and the visualization player is delivered inside the html pages that the client ask at the beginning of the connection. On the server side a software called AudioVideo Engine generates the phonemes and visemes information needed for the animation. The demo called Emotional Parrot shows the ability to reproduce the same input in different emotional states. This is the first WebGL software running on iOS device ever.
See at:
dl.acm.org 
| CNR IRIS | CNR IRIS 
2011
Conference article
Restricted
LUCIA-WebGL: A Web Based Italian MPEG-4 Talking Head
Leone G. R., Cosi P.
In this work we present the reviewing of the activities focused on the development of the WebGL software version of LUCIA talking head, an open source facial animation framework developed at ISTC-CNR of Padua. LUCIA works on standard MPEG-4 Facial Animation Parameters and speaks with the Italian version of FESTIVAL TTS. LUCIA is totally based on true real human data collected by the use of ELITE, a fully automatic movement analyzer for 3D kinematics data acquisition. These informations are used to create lips articulatory model and to drive directly the talking face, generating human facial movements. We are exploiting the use of LUCIA WebGL as a virtual guide in the Wikimemo.it project: The portal of Italian Language and Culture. The easy integration of this technology in websites offers promising future uses for theWebGL Avatars: on-line personal assistant, storyteller for web-books, digital tutor for hearing impaired are only few examples. Index Terms: WebGL, talking head, facial animation, mpeg4, 3D avatar, virtual agent, TTS, LUCIA, FESTIVAL
Leone G. R., Cosi P.
In this work we present the reviewing of the activities focused on the development of the WebGL software version of LUCIA talking head, an open source facial animation framework developed at ISTC-CNR of Padua. LUCIA works on standard MPEG-4 Facial Animation Parameters and speaks with the Italian version of FESTIVAL TTS. LUCIA is totally based on true real human data collected by the use of ELITE, a fully automatic movement analyzer for 3D kinematics data acquisition. These informations are used to create lips articulatory model and to drive directly the talking face, generating human facial movements. We are exploiting the use of LUCIA WebGL as a virtual guide in the Wikimemo.it project: The portal of Italian Language and Culture. The easy integration of this technology in websites offers promising future uses for theWebGL Avatars: on-line personal assistant, storyteller for web-books, digital tutor for hearing impaired are only few examples. Index Terms: WebGL, talking head, facial animation, mpeg4, 3D avatar, virtual agent, TTS, LUCIA, FESTIVAL
2023
Other
Open Access
L'Intelligenza Artificiale nella elaborazione delle immagini: tecniche di visione artificiale per il monitoraggio della guida di motocicli
Bulotta D., Carta A., Righi M., Leone G. R.
Il tirocinio si è svolto presso l'Istituto di Scienza e Tecnologie dell'Informazione del Consiglio Nazionale delle Ricerche e ha richiesto circa cinque mesi. Il tirocinio si è svolto nell'ambito del progetto di ricerca Artificial Intelligence driven RIding Distributed Eye (AI-RIDE) che ha come obiettivo lo studio e la realizzazione prototipale di un framework di Intelligenza Artificiale integrato nel contesto della formazione dei motociclisti, mirando in particolare ad una verifica standard, misurabile e imparziale dell'esame della patente di guida per motocicli. L'esito di un esame di patente di guida dipende da diversi fattori, come il tempo di prestazione, la precisione della traiettoria, la gestione della velocità e l'esecuzione di un percorso netto senza effettuare penalità. La maggior parte di tali fattori può essere misurata utilizzando strumenti di analisi delle immagini ottenute per mezzo di telecamere esterne opportunamente posizionate. L'obiettivo di questo tirocinio è stato ricavare tali informazioni dai flussi video a disposizione massimizzando la precisione dei dati ottenuti. Il primo compito è consistito nell'allenamento di un modello di Object Detection per il riconoscimento dei coni, del motociclo e del pilota; utilizzando vari strumenti di annotazione si è popolato un dataset con numerosi frame estratti dai video realizzati nel sito di prova, la pista della scuola guida Gerardo di Pontedera; tale dataset è stato continuamente migliorato aggiungendo immagini di casistiche specifiche, conducendo test a diverse risoluzioni e utilizzando numerosità crescenti dei modelli di partenza (come spiegato nei paragrafi 3.1.7 e 4.2.8). Raggiunta una soddisfacente prestazione del task di riconoscimento si è implementata la logica funzionale alla individuazione precisa della posizione della moto nello spazio e all'eventuale penalità connessa con lo spostamento di uno o più coni segnaletici (si veda paragrafo 4.3.1 e 4.3.2). Queste informazioni sono servite da input al sottosistema di fusione delle informazioni che non è compreso nel lavoro di questo tirocinio. Il modello custom di object detection ha conseguito ottimi risultati in termini di confidenza e precisione nel riconoscimento degli elementi in gioco ovvero coni, moto e pilota. La logica implementata per il rilevamento delle penalità commesse durante l'esame di guida è stata essenziale per il successo finale del sistema. Il lavoro svolto è stato presentato il giorno 14 giugno 2023 presso il sito di prova in un evento che ha previsto una dimostrazione del sistema dal vivo includendo vari scenari di guida: tutte le penalità principali sono state segnalate con elevata precisione spaziale e temporale.
Bulotta D., Carta A., Righi M., Leone G. R.
Il tirocinio si è svolto presso l'Istituto di Scienza e Tecnologie dell'Informazione del Consiglio Nazionale delle Ricerche e ha richiesto circa cinque mesi. Il tirocinio si è svolto nell'ambito del progetto di ricerca Artificial Intelligence driven RIding Distributed Eye (AI-RIDE) che ha come obiettivo lo studio e la realizzazione prototipale di un framework di Intelligenza Artificiale integrato nel contesto della formazione dei motociclisti, mirando in particolare ad una verifica standard, misurabile e imparziale dell'esame della patente di guida per motocicli. L'esito di un esame di patente di guida dipende da diversi fattori, come il tempo di prestazione, la precisione della traiettoria, la gestione della velocità e l'esecuzione di un percorso netto senza effettuare penalità. La maggior parte di tali fattori può essere misurata utilizzando strumenti di analisi delle immagini ottenute per mezzo di telecamere esterne opportunamente posizionate. L'obiettivo di questo tirocinio è stato ricavare tali informazioni dai flussi video a disposizione massimizzando la precisione dei dati ottenuti. Il primo compito è consistito nell'allenamento di un modello di Object Detection per il riconoscimento dei coni, del motociclo e del pilota; utilizzando vari strumenti di annotazione si è popolato un dataset con numerosi frame estratti dai video realizzati nel sito di prova, la pista della scuola guida Gerardo di Pontedera; tale dataset è stato continuamente migliorato aggiungendo immagini di casistiche specifiche, conducendo test a diverse risoluzioni e utilizzando numerosità crescenti dei modelli di partenza (come spiegato nei paragrafi 3.1.7 e 4.2.8). Raggiunta una soddisfacente prestazione del task di riconoscimento si è implementata la logica funzionale alla individuazione precisa della posizione della moto nello spazio e all'eventuale penalità connessa con lo spostamento di uno o più coni segnaletici (si veda paragrafo 4.3.1 e 4.3.2). Queste informazioni sono servite da input al sottosistema di fusione delle informazioni che non è compreso nel lavoro di questo tirocinio. Il modello custom di object detection ha conseguito ottimi risultati in termini di confidenza e precisione nel riconoscimento degli elementi in gioco ovvero coni, moto e pilota. La logica implementata per il rilevamento delle penalità commesse durante l'esame di guida è stata essenziale per il successo finale del sistema. Il lavoro svolto è stato presentato il giorno 14 giugno 2023 presso il sito di prova in un evento che ha previsto una dimostrazione del sistema dal vivo includendo vari scenari di guida: tutte le penalità principali sono state segnalate con elevata precisione spaziale e temporale.
See at:
ISTI Repository | CNR ExploRA
2023
Other
Open Access
Monitoraggio della guida di motocicli per mezzo di un sistema di visione multicamera con tecniche di fusione delle informazioni e loro rappresentazione virtuale per mezzo di un motore grafico 3D
Baiamonte A., Bacciu D., Righi M., Leone G. R.
Il tirocinio si è svolto presso l'istituto di Scienza e Tecnologie dell'informazione del Consiglio Nazionale delle Ricerche (ISTI-CNR), da metà Marzo a fine Giugno 2023. L'obiettivo del progetto è stato il monitoraggio della guida di motocicli per mezzo di un sistema di visione multicamera con tecniche di fusione delle informazioni e loro rappresentazione virtuale per mezzo di un motore grafico 3D. Il tirocinio è stato realizzato nell'ambito di un progetto di ricerca del CNR dedicato alla progetta- zione e prototipazione di un sistema multicamera innovativo [1], basato su algoritmi di Computer Vision e Intelligenza Artificiale, in grado di riconoscere automaticamente, con grado di affidabilità variabile, le penalità previste durante la fase pratica degli esami per il conseguimento della patente di guida per motocicli. I dati raccolti dai singoli flussi video sono stati elaborati e fusi per ottenere un'informazione globale più completa ed affidabile. Particolare attenzione è stata dedicata al calcolo vettoriale delle accelerazioni del motoveicolo nei test in circuito chiuso. Inoltre, è stato realizzato un Digital Twin (gemello virtuale) del sistema per visualizzare in grafica 3D i dati elaborati e permettere l'analisi delle traiettorie effettuate dal motoveicolo in un contesto di prove ripetute. Nel capitolo 2 si esporranno le basi di partenza del tirocinio, ovvero cosa era già presente prima di iniziare questo lavoro di tesi. Nel capitolo 3 si parlerà del background necessario a descrivere le metodologie e le tecnologie utilizzate in questa tesi. Nel capitolo 4 si elencheranno i software utilizzati per realizzare le varie funzionalità del sistema mettendo in evidenza le scelte effettuate, giustificando perché è stata scelta una particolare tecnologia piuttosto che un'altra. Nel capitolo 5 si descriverà il lavoro svolto durante il tirocinio mostrando tutti i test, gli esperimenti fatti e risultati ottenuti. Il capitolo 6 conclude la tesi riassumendo quanto si è realizzato e parlando dei possibili sviluppi futuri.
Baiamonte A., Bacciu D., Righi M., Leone G. R.
Il tirocinio si è svolto presso l'istituto di Scienza e Tecnologie dell'informazione del Consiglio Nazionale delle Ricerche (ISTI-CNR), da metà Marzo a fine Giugno 2023. L'obiettivo del progetto è stato il monitoraggio della guida di motocicli per mezzo di un sistema di visione multicamera con tecniche di fusione delle informazioni e loro rappresentazione virtuale per mezzo di un motore grafico 3D. Il tirocinio è stato realizzato nell'ambito di un progetto di ricerca del CNR dedicato alla progetta- zione e prototipazione di un sistema multicamera innovativo [1], basato su algoritmi di Computer Vision e Intelligenza Artificiale, in grado di riconoscere automaticamente, con grado di affidabilità variabile, le penalità previste durante la fase pratica degli esami per il conseguimento della patente di guida per motocicli. I dati raccolti dai singoli flussi video sono stati elaborati e fusi per ottenere un'informazione globale più completa ed affidabile. Particolare attenzione è stata dedicata al calcolo vettoriale delle accelerazioni del motoveicolo nei test in circuito chiuso. Inoltre, è stato realizzato un Digital Twin (gemello virtuale) del sistema per visualizzare in grafica 3D i dati elaborati e permettere l'analisi delle traiettorie effettuate dal motoveicolo in un contesto di prove ripetute. Nel capitolo 2 si esporranno le basi di partenza del tirocinio, ovvero cosa era già presente prima di iniziare questo lavoro di tesi. Nel capitolo 3 si parlerà del background necessario a descrivere le metodologie e le tecnologie utilizzate in questa tesi. Nel capitolo 4 si elencheranno i software utilizzati per realizzare le varie funzionalità del sistema mettendo in evidenza le scelte effettuate, giustificando perché è stata scelta una particolare tecnologia piuttosto che un'altra. Nel capitolo 5 si descriverà il lavoro svolto durante il tirocinio mostrando tutti i test, gli esperimenti fatti e risultati ottenuti. Il capitolo 6 conclude la tesi riassumendo quanto si è realizzato e parlando dei possibili sviluppi futuri.
See at:
ISTI Repository | CNR ExploRA
2017
Journal article
Open Access
An intelligent cooperative visual sensor network for urban mobility
Leone G R, Moroni D, Pieri G, Petracca M, Salvetti O, Azzarà A, Marino F
Smart cities are demanding solutions for improved traffic efficiency, in order to guarantee optimal access to mobility resources available in urban areas. Intelligent video analytics deployed directly on board embedded sensors offers great opportunities to gather highly informative data about traffic and transport, allowing reconstruction of a real-time neat picture of urban mobility patterns. In this paper, we present a visual sensor network in which each node embeds computer vision logics for analyzing in real time urban traffic. The nodes in the network share their perceptions and build a global and comprehensive interpretation of the analyzed scenes in a cooperative and adaptive fashion. This is possible thanks to an especially designed Internet of Things (IoT) compliant middleware which encompasses in-network event composition as well as full support of Machine-2-Machine (M2M) communication mechanism. The potential of the proposed cooperative visual sensor network is shown with two sample applications in urban mobility connected to the estimation of vehicular flows and parking management. Besides providing detailed results of each key component of the proposed solution, the validity of the approach is demonstrated by extensive field tests that proved the suitability of the system in providing a scalable, adaptable and extensible data collection layer for managing and understanding mobility in smart cities.Source: SENSORS (BASEL), vol. 17 (issue 11)
DOI: 10.3390/s17112588
Project(s): ICSI
Metrics:
Leone G R, Moroni D, Pieri G, Petracca M, Salvetti O, Azzarà A, Marino F
Smart cities are demanding solutions for improved traffic efficiency, in order to guarantee optimal access to mobility resources available in urban areas. Intelligent video analytics deployed directly on board embedded sensors offers great opportunities to gather highly informative data about traffic and transport, allowing reconstruction of a real-time neat picture of urban mobility patterns. In this paper, we present a visual sensor network in which each node embeds computer vision logics for analyzing in real time urban traffic. The nodes in the network share their perceptions and build a global and comprehensive interpretation of the analyzed scenes in a cooperative and adaptive fashion. This is possible thanks to an especially designed Internet of Things (IoT) compliant middleware which encompasses in-network event composition as well as full support of Machine-2-Machine (M2M) communication mechanism. The potential of the proposed cooperative visual sensor network is shown with two sample applications in urban mobility connected to the estimation of vehicular flows and parking management. Besides providing detailed results of each key component of the proposed solution, the validity of the approach is demonstrated by extensive field tests that proved the suitability of the system in providing a scalable, adaptable and extensible data collection layer for managing and understanding mobility in smart cities.Source: SENSORS (BASEL), vol. 17 (issue 11)
DOI: 10.3390/s17112588
Project(s): ICSI
Metrics:
See at:
Sensors | Sensors | CNR IRIS | ISTI Repository | Sensors | Sensors | ZENODO | Hyper Article en Ligne | CNR IRIS
2018
Journal article
Open Access
Real-time smart parking systems integration in distributed ITS for smart cities
Alam M, Moroni D, Pieri G, Tampucci M, Gomes M, Fonseca J, Ferreira J, Leone Gr
Intelligent Transportation Systems (ITS) have evolved as a key research topic in recent years, revolutionizing the overall traffic and travel experience by providing a set of advanced services and applications. These data-driven services contribute to mitigate major problems arising from the ever growing need of transport in our daily lives. Despite the progress, there is still need for an enhanced and distributed solution that can exploit the data from the available systems and provide an appropriate and real-time reaction on transportation systems. Therefore, in this paper, we present a new architecture where the intelligence is distributed and the decisions are decentralized. The proposed architecture is scalable since the incremental addition of new peripheral subsystems is supported by the introduction of gateways which requires no reengineering of the communication infrastructure. The proposed architecture is deployed to tackle the problem of traffic management inefficiency in urban areas, where traffic load is substantially increased, by vehicles moving around unnecessarily, to find a free parking space. This can be significantly reduced through the availability and diffusion of local information regarding vacant parking slots to drivers in a given area. Two types of parking systems, magnetic and vision sensor based, have been introduced, deployed, and tested in different scenarios. The effectiveness of the proposed architecture, together with the proposed algorithms, is assessed in field trials.Source: JOURNAL OF ADVANCED TRANSPORTATION (ONLINE), vol. 2018 (issue 3437278)
Project(s): ICSI
Alam M, Moroni D, Pieri G, Tampucci M, Gomes M, Fonseca J, Ferreira J, Leone Gr
Intelligent Transportation Systems (ITS) have evolved as a key research topic in recent years, revolutionizing the overall traffic and travel experience by providing a set of advanced services and applications. These data-driven services contribute to mitigate major problems arising from the ever growing need of transport in our daily lives. Despite the progress, there is still need for an enhanced and distributed solution that can exploit the data from the available systems and provide an appropriate and real-time reaction on transportation systems. Therefore, in this paper, we present a new architecture where the intelligence is distributed and the decisions are decentralized. The proposed architecture is scalable since the incremental addition of new peripheral subsystems is supported by the introduction of gateways which requires no reengineering of the communication infrastructure. The proposed architecture is deployed to tackle the problem of traffic management inefficiency in urban areas, where traffic load is substantially increased, by vehicles moving around unnecessarily, to find a free parking space. This can be significantly reduced through the availability and diffusion of local information regarding vacant parking slots to drivers in a given area. Two types of parking systems, magnetic and vision sensor based, have been introduced, deployed, and tested in different scenarios. The effectiveness of the proposed architecture, together with the proposed algorithms, is assessed in field trials.Source: JOURNAL OF ADVANCED TRANSPORTATION (ONLINE), vol. 2018 (issue 3437278)
Project(s): ICSI
See at:
CNR IRIS | ISTI Repository | www.hindawi.com | CNR IRIS
2019
Journal article
Open Access
Fault detection in power equipment via an unmanned aerial system using multi modal data
Jalil B, Leone Gr, Martinelli M, Moroni M, Pascali Ma, Berton A
The power transmission lines are the link between the power plants and the points of consumption, through substations. Most importantly, the assessment of damaged aerial power lines and rusted conductors is of extreme importance for public safety; hence, power line and associated components must be periodically inspected to ensure a continuous supply and to identify any fault and defect. To achieve these objectives, recently Unmanned Aerial Vehicles (UAVs) have been widely used: in fact, they provide a safe way to bring sensors close to the power transmission lines and their associated components without halting the equipment during the inspection, and reducing operational cost and risk. In this work, the drone, equipped with multi-modal sensors, captures images in the visible and infrared domain and transmits them to the ground station. We used state-of-the-art computer vision methods to highlight expected faults (i.e. hot spots) or damaged components of the electrical infrastructure (i.e. damaged insulators). Infrared imaging, which is invariant to large scale and illumination changes in the real operating environment, supported the identification of faults in power transmission lines; while a neural network is adapted and trained to detect and classify insulators from an optical video stream. We demonstrate our approach on the data captured by a drone in Parma, Italy.Source: SENSORS (BASEL), vol. 19 (issue 13)
Jalil B, Leone Gr, Martinelli M, Moroni M, Pascali Ma, Berton A
The power transmission lines are the link between the power plants and the points of consumption, through substations. Most importantly, the assessment of damaged aerial power lines and rusted conductors is of extreme importance for public safety; hence, power line and associated components must be periodically inspected to ensure a continuous supply and to identify any fault and defect. To achieve these objectives, recently Unmanned Aerial Vehicles (UAVs) have been widely used: in fact, they provide a safe way to bring sensors close to the power transmission lines and their associated components without halting the equipment during the inspection, and reducing operational cost and risk. In this work, the drone, equipped with multi-modal sensors, captures images in the visible and infrared domain and transmits them to the ground station. We used state-of-the-art computer vision methods to highlight expected faults (i.e. hot spots) or damaged components of the electrical infrastructure (i.e. damaged insulators). Infrared imaging, which is invariant to large scale and illumination changes in the real operating environment, supported the identification of faults in power transmission lines; while a neural network is adapted and trained to detect and classify insulators from an optical video stream. We demonstrate our approach on the data captured by a drone in Parma, Italy.Source: SENSORS (BASEL), vol. 19 (issue 13)
See at:
CNR IRIS | ISTI Repository | ISTI Repository | www.mdpi.com | CNR IRIS | CNR IRIS
2019
Conference article
Open Access
Smart cities monitoring through wireless smart cameras
Moroni D, Pieri G, Leone Gr, Tampucci M
Cities and urban environments nowadays suffer more and more discomfort due to several factors: the increase in the number of cars, the reduction, due to budget issues, of free parking, the economic difficulties in times of crisis that lead to a reduced efficiency of the public transport system. Beside these factors, an increased availability of computing facilities and devices bring the possibility to improve one of the most burdening aspects, that is the traffic generated by cars searching for a parking place. In this paper we present an approach based on wireless smart cameras to the monitoring of open air public parking areas. The main strengths of this solution lie in the independence and scalability of the proposed architecture. Independence is mainly based on autonomous intelligent cameras performing processing on-board; while scalability is mainly obtained through low cost single node which can be assembled in a wider sensors network and do not need high-cost requirements for installation.
Moroni D, Pieri G, Leone Gr, Tampucci M
Cities and urban environments nowadays suffer more and more discomfort due to several factors: the increase in the number of cars, the reduction, due to budget issues, of free parking, the economic difficulties in times of crisis that lead to a reduced efficiency of the public transport system. Beside these factors, an increased availability of computing facilities and devices bring the possibility to improve one of the most burdening aspects, that is the traffic generated by cars searching for a parking place. In this paper we present an approach based on wireless smart cameras to the monitoring of open air public parking areas. The main strengths of this solution lie in the independence and scalability of the proposed architecture. Independence is mainly based on autonomous intelligent cameras performing processing on-board; while scalability is mainly obtained through low cost single node which can be assembled in a wider sensors network and do not need high-cost requirements for installation.
See at:
dl.acm.org | CNR IRIS | ISTI Repository | CNR IRIS | CNR IRIS
2019
Conference article
Open Access
Smart Cities: parking monitoring through smart cameras
Leone Gr, Moroni D, Pieri G
One of today's major problems in medium to large cities is pollution caused by urban traffic. There are various factors that negatively affect this problem. First of all, the constant increase in vehicles, and at the same time the reduction, for budget reasons, of the number of free parking areas more or less easily reached. Moreover, due to global events such as the economic crisis, there is a general reduction in the efficiency of local public transport. In this research, we present an approach to this problem that exploits the greater availability of IT structures and devices. The solution presented offers the possibility of improving one of the most burdensome aspects, namely the traffic generated by cars looking for a parking space through an integrated and automatic management of employment levels. The procedure is based on wireless smart cameras for monitoring public outdoor parking areas. The main strengths of this solution lie in the independence and scalability of the proposed architecture. Independence is ensured by the node's modus operandi, in which processing is performed on board the intelligent camera without the need for a video stream or operator intervention. Scalability can be achieved by designing low-cost nodes that can be connected in a sensor network that does not require high installation costs. Moreover, the possibility of making such processing nodes autonomous through the use of a photovoltaic panel allows them to be installed in even more flimsy situations.
Leone Gr, Moroni D, Pieri G
One of today's major problems in medium to large cities is pollution caused by urban traffic. There are various factors that negatively affect this problem. First of all, the constant increase in vehicles, and at the same time the reduction, for budget reasons, of the number of free parking areas more or less easily reached. Moreover, due to global events such as the economic crisis, there is a general reduction in the efficiency of local public transport. In this research, we present an approach to this problem that exploits the greater availability of IT structures and devices. The solution presented offers the possibility of improving one of the most burdensome aspects, namely the traffic generated by cars looking for a parking space through an integrated and automatic management of employment levels. The procedure is based on wireless smart cameras for monitoring public outdoor parking areas. The main strengths of this solution lie in the independence and scalability of the proposed architecture. Independence is ensured by the node's modus operandi, in which processing is performed on board the intelligent camera without the need for a video stream or operator intervention. Scalability can be achieved by designing low-cost nodes that can be connected in a sensor network that does not require high installation costs. Moreover, the possibility of making such processing nodes autonomous through the use of a photovoltaic panel allows them to be installed in even more flimsy situations.
See at:
CNR IRIS | ieeexplore.ieee.org | ISTI Repository | CNR IRIS
2018
Journal article
Open Access
Corrigendum to "Real-Time Smart Parking Systems Integration in Distributed ITS for Smart Cities"
Alam M, Moroni D, Pieri G, Tampucci M, Gomes M, Fonseca J, Ferreira J, Leone Gr
In the article titled "Real-Time Smart Parking Systems Integration in Distributed ITS for Smart Cities", Dr. Giuseppe Riccardo Leone was missing from the authors' list. Dr. Leone partially contributed in the designing of the study and mainly in the preparation, programming, and execution of the algorithms regarding the field trials and tests. He contributed in obtaining and analyzing the data. The corrected authors' list is shown above and updated in place.Source: JOURNAL OF ADVANCED TRANSPORTATION (ONLINE)
Project(s): ICSI
Alam M, Moroni D, Pieri G, Tampucci M, Gomes M, Fonseca J, Ferreira J, Leone Gr
In the article titled "Real-Time Smart Parking Systems Integration in Distributed ITS for Smart Cities", Dr. Giuseppe Riccardo Leone was missing from the authors' list. Dr. Leone partially contributed in the designing of the study and mainly in the preparation, programming, and execution of the algorithms regarding the field trials and tests. He contributed in obtaining and analyzing the data. The corrected authors' list is shown above and updated in place.Source: JOURNAL OF ADVANCED TRANSPORTATION (ONLINE)
Project(s): ICSI
See at:
CNR IRIS | ISTI Repository | www.hindawi.com | CNR IRIS
2022
Conference article
Open Access
Computer vision per sistemi di trasporto intelligenti: il progetto S.Pa.Ce.
Leone Gr, Carboni A, Nardi S, Moroni D
Lo Smart Passenger Center (SPaCe) è una piattaforma integrata che mira a superare la complessità della gestione centralizzata delle infrastrutture di trasporto pubblico e dei veicoli. Il motore di intelligenza artificiale esamina i flussi quotidiani di persone, correla dati ed eventi diversi, prevede minacce ed eventi critici e propone contromisure. Questa enorme mole di dati proviene da una rete pervasiva di telecamere intelligenti che monitora costantemente le attività in stazioni, treni, autobus e altri luoghi di interesse. In questo lavoro, presentiamo il sottosistema distribuito di visione artificiale, lo stato dell'arte delle tecniche adottate e le funzionalità avanzate che questo sistema di sorveglianza intelligente offre ai livelli superiori di SPaCe. Tutto è sviluppato seguendo il paradigma della privacy-by-design: nessuna immagine viene registrata o trasmessa, ma tutte le elaborazioni avvengono sui nodi periferici del sistema.
Leone Gr, Carboni A, Nardi S, Moroni D
Lo Smart Passenger Center (SPaCe) è una piattaforma integrata che mira a superare la complessità della gestione centralizzata delle infrastrutture di trasporto pubblico e dei veicoli. Il motore di intelligenza artificiale esamina i flussi quotidiani di persone, correla dati ed eventi diversi, prevede minacce ed eventi critici e propone contromisure. Questa enorme mole di dati proviene da una rete pervasiva di telecamere intelligenti che monitora costantemente le attività in stazioni, treni, autobus e altri luoghi di interesse. In questo lavoro, presentiamo il sottosistema distribuito di visione artificiale, lo stato dell'arte delle tecniche adottate e le funzionalità avanzate che questo sistema di sorveglianza intelligente offre ai livelli superiori di SPaCe. Tutto è sviluppato seguendo il paradigma della privacy-by-design: nessuna immagine viene registrata o trasmessa, ma tutte le elaborazioni avvengono sui nodi periferici del sistema.
See at:
CNR IRIS | ISTI Repository | ISTI Repository | www.ital-ia2022.it | CNR IRIS | CNR IRIS
2022
Conference article
Open Access
Toward pervasive computer vision for intelligent transport system
Leone Gr, Carboni A, Nardi S, Moroni D
The Smart Passenger Center (SPaCe) is a fully integrated platform that aims to overcome the complexity of centralized management of public transport infrastructure and vehicles. The SPaCe artificial intelligence engine predicts threats and critical events and proposes countermeasures by examining the daily flows of people and correlating different data and events, thanks to machine learning and big data analytics. All this massive data comes from a pervasive smart cameras network that constantly monitors activities in stations, trains, buses and other places of interest. In this work, we present the idea of this computer vision distributed sub-system, the state of the art of the techniques involved and the advanced functionalities that this intelligent surveillance system offers to the upper layers. Everything is developed following the privacy-by-design paradigm; namely, no real image is recorded or transmitted, but all the elaborations take place on the edge nodes of the system.
Leone Gr, Carboni A, Nardi S, Moroni D
The Smart Passenger Center (SPaCe) is a fully integrated platform that aims to overcome the complexity of centralized management of public transport infrastructure and vehicles. The SPaCe artificial intelligence engine predicts threats and critical events and proposes countermeasures by examining the daily flows of people and correlating different data and events, thanks to machine learning and big data analytics. All this massive data comes from a pervasive smart cameras network that constantly monitors activities in stations, trains, buses and other places of interest. In this work, we present the idea of this computer vision distributed sub-system, the state of the art of the techniques involved and the advanced functionalities that this intelligent surveillance system offers to the upper layers. Everything is developed following the privacy-by-design paradigm; namely, no real image is recorded or transmitted, but all the elaborations take place on the edge nodes of the system.
See at:
CNR IRIS | ieeexplore.ieee.org | ISTI Repository | CNR IRIS | CNR IRIS
2022
Conference article
Open Access
Towards multi-camera system for the evaluation of motorcycle driving test
Leone Gr, Righi M, Moroni D, Paolucci F
This work describes the early stage of an interactive and accelerated AI-driven framework for Practical Driving Courses and Driving Licence Exams. The core of the project is an innovative multi-parameter AI-assisted telemetry system able to compute test scores and outcome, useful for human-neutral auditability of Driving Licence Exams. The distributed Artificial Intelligence (AI) system available at the Track Testbed will be able to perform driving behaviour classifications and will suggest specific improvements based on the analysis of vehicle trajectories acquired during the driving test. Finally, the project will target the creation of a large dataset for driving test classification of key performance parameters. The system is envisioned to have a relevant impact on all the certification, driving licence operators and regulator entities.
Leone Gr, Righi M, Moroni D, Paolucci F
This work describes the early stage of an interactive and accelerated AI-driven framework for Practical Driving Courses and Driving Licence Exams. The core of the project is an innovative multi-parameter AI-assisted telemetry system able to compute test scores and outcome, useful for human-neutral auditability of Driving Licence Exams. The distributed Artificial Intelligence (AI) system available at the Track Testbed will be able to perform driving behaviour classifications and will suggest specific improvements based on the analysis of vehicle trajectories acquired during the driving test. Finally, the project will target the creation of a large dataset for driving test classification of key performance parameters. The system is envisioned to have a relevant impact on all the certification, driving licence operators and regulator entities.
See at:
CNR IRIS | ieeexplore.ieee.org | ISTI Repository | CNR IRIS | CNR IRIS
2023
Conference article
Open Access
Medical waste sorting: a computer vision approach for assisted primary sorting
Bruno A, Caudai C, Leone Gr, Martinelli M, Moroni D, Crotti F
Medical waste, i.e. waste produced during medical activities in hospitals, clinics and laboratories, represents hazardous waste whose management requires special care and high costs. However, this kind of waste contains a large fraction of highly valued materials that can enter a circular economy process. To this end, in this paper, we propose a computer vision approach for assisting in the primary sorting of med- ical waste. The feasibility of our approach is demonstrated on representative datasets we collected and made available to the community.
Bruno A, Caudai C, Leone Gr, Martinelli M, Moroni D, Crotti F
Medical waste, i.e. waste produced during medical activities in hospitals, clinics and laboratories, represents hazardous waste whose management requires special care and high costs. However, this kind of waste contains a large fraction of highly valued materials that can enter a circular economy process. To this end, in this paper, we propose a computer vision approach for assisting in the primary sorting of med- ical waste. The feasibility of our approach is demonstrated on representative datasets we collected and made available to the community.
See at:
CNR IRIS | ieeexplore.ieee.org | ISTI Repository | CNR IRIS | CNR IRIS
2023
Conference article
Open Access
A novel smart camera network for real time public transport monitoring and surveillance
Carboni A., Leone G. R., Nardi S., Corrado A., Moroni D.
In this paper, we present the environment perception layer of the Smart Passenger Center (SPaCe), a novel integrated framework for public transport management. This layer is a pervasive vision architecture for improved safety and security in the context of public transportation. Privacy and technological constraints are still significant limitations for the real-time analysis of video streams from video capture devices installed on public transport vehicles. In fact, in almost all cases, this analysis is carried out offline and lacks any predictive processing, which is now potentially applicable to all transport sectors, thanks to machine learning and artificial vision techniques. The architecture described is designed to combine the output of a set of parallel processing, all running onboard in real-time, thus allowing the separation of the information collected from actual passengers' identities. The analysis highlights aspects that affect travel and travellers safety, such as people's behaviour and the state of maintenance of vehicles.Source: PROCEEDINGS IEEE CONFERENCE ON INTELLIGENT TRANSPORTATION SYSTEMS, pp. 5223-5228. Bilbao, Spain, 24-28/09/2023
Carboni A., Leone G. R., Nardi S., Corrado A., Moroni D.
In this paper, we present the environment perception layer of the Smart Passenger Center (SPaCe), a novel integrated framework for public transport management. This layer is a pervasive vision architecture for improved safety and security in the context of public transportation. Privacy and technological constraints are still significant limitations for the real-time analysis of video streams from video capture devices installed on public transport vehicles. In fact, in almost all cases, this analysis is carried out offline and lacks any predictive processing, which is now potentially applicable to all transport sectors, thanks to machine learning and artificial vision techniques. The architecture described is designed to combine the output of a set of parallel processing, all running onboard in real-time, thus allowing the separation of the information collected from actual passengers' identities. The analysis highlights aspects that affect travel and travellers safety, such as people's behaviour and the state of maintenance of vehicles.Source: PROCEEDINGS IEEE CONFERENCE ON INTELLIGENT TRANSPORTATION SYSTEMS, pp. 5223-5228. Bilbao, Spain, 24-28/09/2023
See at:
CNR IRIS | ieeexplore.ieee.org | CNR IRIS | CNR IRIS
2023
Conference article
Open Access
AI-RIDE: a multi-camera system for the evaluation of motorcycle driving test
Leone Gr, Righi M, Moroni D, Baiamonte A, Bulotta D, Paolucci F
The AI-RIDE project proposes adopting an accelerated, online, and embedded Artificial Intelligence framework in motorcycle rider training, mainly targeting the Practical Driving Courses (PDC) and Driving License Exam (DLE) sessions verification tools. The project targets a disruptive innovation step in the context of driving learning techniques, significantly going beyond the state of the art of the current instruments used in the PDC and DLE ecosystem. This work presents last year's activities with the promising results obtained with the first working prototype.
Leone Gr, Righi M, Moroni D, Baiamonte A, Bulotta D, Paolucci F
The AI-RIDE project proposes adopting an accelerated, online, and embedded Artificial Intelligence framework in motorcycle rider training, mainly targeting the Practical Driving Courses (PDC) and Driving License Exam (DLE) sessions verification tools. The project targets a disruptive innovation step in the context of driving learning techniques, significantly going beyond the state of the art of the current instruments used in the PDC and DLE ecosystem. This work presents last year's activities with the promising results obtained with the first working prototype.
See at:
CNR IRIS | ieeexplore.ieee.org | ISTI Repository | CNR IRIS | CNR IRIS
2024
Journal article
Open Access
Transforming public transport through responsible AI: insight from the FAITH project
Leone G. R., Carboni A., Moroni D., Colantonio S.
In the framework of the FAITH project, an AI-powered large-scale pilot is being organized to showcase the opportunities offered by advancements in edge computing and computer vision. The goal is to provide innovative ways to ensure safer and more sustainable public transport, encouraging modal shifts. Challenges in deploying responsible and trustworthy AI-based services will be addressed by assessing regulatory aspects and fostering co-design with end users and all the other stakeholders involved.Source: ERCIM NEWS, vol. 138, pp. 23-24
Project(s): FAITH
Leone G. R., Carboni A., Moroni D., Colantonio S.
In the framework of the FAITH project, an AI-powered large-scale pilot is being organized to showcase the opportunities offered by advancements in edge computing and computer vision. The goal is to provide innovative ways to ensure safer and more sustainable public transport, encouraging modal shifts. Challenges in deploying responsible and trustworthy AI-based services will be addressed by assessing regulatory aspects and fostering co-design with end users and all the other stakeholders involved.Source: ERCIM NEWS, vol. 138, pp. 23-24
Project(s): FAITH
See at:
ercim-news.ercim.eu | CNR IRIS | CNR IRIS
2024
Conference article
Open Access
AI-RIDE: un sistema multicamera per la valutazione automatica della guida dei motocicli
Leone G. R., Righi M., Moroni D., Bulotta D., Baiamonte A., Paolucci F.
In questo articolo viene presentato il prototipo AI-RIDE, un sistema di visione multicamera per il monitoraggio e la valutazione della guida di motocicli. Per mezzo di algoritmi di Computer Vision e di Intelligenza Artificiale, il sistema è in grado di riconoscere automaticamente, con grado di affidabilità variabile, le penalità previste durante la fase pratica degli esami per il conseguimento della patente di guida per motocicli. Mediante l’analisi dei flussi video e l’elaborazione dei dati ottenuti, si rilevano in modo automatico gli errori commessi e viene assegnato un punteggio alla guida del pilota. Il calcolo del punteggio, una misura oggettiva dell’esito della guida, dipende da diversi fattori: rispetto delle tempistiche, precisione delle traiettorie e in generale padronanza nella gestione del motociclo.
Leone G. R., Righi M., Moroni D., Bulotta D., Baiamonte A., Paolucci F.
In questo articolo viene presentato il prototipo AI-RIDE, un sistema di visione multicamera per il monitoraggio e la valutazione della guida di motocicli. Per mezzo di algoritmi di Computer Vision e di Intelligenza Artificiale, il sistema è in grado di riconoscere automaticamente, con grado di affidabilità variabile, le penalità previste durante la fase pratica degli esami per il conseguimento della patente di guida per motocicli. Mediante l’analisi dei flussi video e l’elaborazione dei dati ottenuti, si rilevano in modo automatico gli errori commessi e viene assegnato un punteggio alla guida del pilota. Il calcolo del punteggio, una misura oggettiva dell’esito della guida, dipende da diversi fattori: rispetto delle tempistiche, precisione delle traiettorie e in generale padronanza nella gestione del motociclo.
See at:
CNR IRIS | ital-ia2024.it | CNR IRIS
2018
Conference article
Open Access
Towards structural monitoring and 3D documentation of architectural heritage using UAV
Germanese D, Leone Gr, Moroni D, Pascali Ma, Tampucci M
This paper describes how UAVs may support the architectural heritage preservation and dissemination. In detail, this work deals with the long-term monitoring of the crack pattern of historic structures, and with the reconstruction of interactive 3D scene in order to provide both the scholar and the general public with a simple and engaging tool to analyze or visit the historic structure.Source: ADVANCES IN INTELLIGENT SYSTEMS AND COMPUTING, vol. 833, pp. 332-342. Wrocraw, Poland, 12-14 September 2018
Germanese D, Leone Gr, Moroni D, Pascali Ma, Tampucci M
This paper describes how UAVs may support the architectural heritage preservation and dissemination. In detail, this work deals with the long-term monitoring of the crack pattern of historic structures, and with the reconstruction of interactive 3D scene in order to provide both the scholar and the general public with a simple and engaging tool to analyze or visit the historic structure.Source: ADVANCES IN INTELLIGENT SYSTEMS AND COMPUTING, vol. 833, pp. 332-342. Wrocraw, Poland, 12-14 September 2018
See at:
CNR IRIS | link.springer.com | ISTI Repository | CNR IRIS | CNR IRIS