AHCI RESEARCH GROUP
Publications
Papers published in international journals,
proceedings of conferences, workshops and books.
OUR RESEARCH
Scientific Publications
How to
Here you can find the complete list of our publications.
You can use the tag cloud to select only the papers dealing with specific research topics.
You can expand the Abstract, Links and BibTex record for each paper.
You can use the tag cloud to select only the papers dealing with specific research topics.
You can expand the Abstract, Links and BibTex record for each paper.
2019
Essmaeel, Kyis; Migniot, Cyrille; Dipanda, Albert; Gallo, Luigi; Damiani, Ernesto; Pietro, Giuseppe De
A New 3D Descriptor for Human Classification: Application for Human Detection in a Multi-Kinect System Journal Article
In: Multimedia Tools and Applications, vol. 78, no. 16, pp. 22479–22508, 2019, ISSN: 1573-7721.
Abstract | Links | BibTeX | Tags: 3D descriptor, Classification, Human detection, Kinect
@article{essmaeelNew3DDescriptor2019,
title = {A New 3D Descriptor for Human Classification: Application for Human Detection in a Multi-Kinect System},
author = { Kyis Essmaeel and Cyrille Migniot and Albert Dipanda and Luigi Gallo and Ernesto Damiani and Giuseppe De Pietro},
doi = {10.1007/s11042-019-7568-6},
issn = {1573-7721},
year = {2019},
date = {2019-08-01},
journal = {Multimedia Tools and Applications},
volume = {78},
number = {16},
pages = {22479--22508},
abstract = {In this paper we present a new 3D descriptor for human classification and a human detection method based on this descriptor. The proposed 3D descriptor allows classification of an object represented by a point cloud, as human or non-human. It is derived from the well-known Histogram of Oriented Gradient by employing surface normals instead of gradients. The process consists in an appropriate subdivision of the object point cloud into blocks. These blocks provide the spatial distribution modeling of the surface normal orientation into the different parts of the object. This distribution modelling is expressed as a histogram. In addition we have set up a multi-kinect acquisition system that provides us with Complete Point Clouds (CPC) (i.e. 360textdegree view). Such CPCs enable a suitable processing, particularly in case of occlusions. Moreover they allow for the determination of the human frontal orientation. Based on the proposed 3D descriptor, we have developed a human detection method that is applied on CPCs. First, we evaluated the 3D descriptor over a set of CPC candidates by using the Support Vector Machine (SVM) classifier. The learning process was conducted with the original CPC database that we have built. The results are very promising. The descriptor can discriminate human from non-human candidates and provides the frontal direction of humans with high precision. In addition we demonstrated that using the CPCs improves significantly the classification results in comparison with Single Point Clouds (i.e. points clouds acquired with only one kinect). Second, we compared our detection method with two others, namely the HOG detector on RGB images and a 3D HOG-based detection method that is applied on RGB-depth data. The obtained results on different situations show that the proposed human detection method provides excellent performances that outperform the other two detection methods.},
keywords = {3D descriptor, Classification, Human detection, Kinect},
pubstate = {published},
tppubtype = {article}
}
In this paper we present a new 3D descriptor for human classification and a human detection method based on this descriptor. The proposed 3D descriptor allows classification of an object represented by a point cloud, as human or non-human. It is derived from the well-known Histogram of Oriented Gradient by employing surface normals instead of gradients. The process consists in an appropriate subdivision of the object point cloud into blocks. These blocks provide the spatial distribution modeling of the surface normal orientation into the different parts of the object. This distribution modelling is expressed as a histogram. In addition we have set up a multi-kinect acquisition system that provides us with Complete Point Clouds (CPC) (i.e. 360textdegree view). Such CPCs enable a suitable processing, particularly in case of occlusions. Moreover they allow for the determination of the human frontal orientation. Based on the proposed 3D descriptor, we have developed a human detection method that is applied on CPCs. First, we evaluated the 3D descriptor over a set of CPC candidates by using the Support Vector Machine (SVM) classifier. The learning process was conducted with the original CPC database that we have built. The results are very promising. The descriptor can discriminate human from non-human candidates and provides the frontal direction of humans with high precision. In addition we demonstrated that using the CPCs improves significantly the classification results in comparison with Single Point Clouds (i.e. points clouds acquired with only one kinect). Second, we compared our detection method with two others, namely the HOG detector on RGB images and a 3D HOG-based detection method that is applied on RGB-depth data. The obtained results on different situations show that the proposed human detection method provides excellent performances that outperform the other two detection methods.
2015
Essmaeel, Kyis; Gallo, Luigi; Damiani, Ernesto; Pietro, Giuseppe De; Dipanda, Albert
Comparative Evaluation of Methods for Filtering Kinect Depth Data Journal Article
In: Multimedia Tools and Applications, vol. 74, no. 17, pp. 7331–7354, 2015, ISSN: 1380-7501.
Abstract | Links | BibTeX | Tags: Comparative evaluation, Depth data, Depth instability, Filtering, Kinect, Median filter, Temporal denoising
@article{essmaeelComparativeEvaluationMethods2015,
title = {Comparative Evaluation of Methods for Filtering Kinect Depth Data},
author = { Kyis Essmaeel and Luigi Gallo and Ernesto Damiani and Giuseppe De Pietro and Albert Dipanda},
doi = {10.1007/s11042-014-1982-6},
issn = {1380-7501},
year = {2015},
date = {2015-09-01},
journal = {Multimedia Tools and Applications},
volume = {74},
number = {17},
pages = {7331--7354},
abstract = {The release of the Kinect has fostered the design of novel methods and techniques in several application domains. It has been tested in different contexts, which span from home entertainment to surgical environments. Nonetheless, to promote its adoption to solve real-world problems, the Kinect should be evaluated in terms of precision and accuracy. Up to now, some filtering approaches have been proposed to enhance the precision and accuracy of the Kinect sensor, and preliminary studies have shown promising results. In this work, we discuss the results of a study in which we have compared the most commonly used filtering approaches for Kinect depth data, in both static and dynamic contexts, by using novel metrics. The experimental results show that each approach can be profitably used to enhance the precision and/or accuracy of Kinect depth data in a specific context, whereas the temporal filtering approach is able to reduce noise in different experimental conditions.},
keywords = {Comparative evaluation, Depth data, Depth instability, Filtering, Kinect, Median filter, Temporal denoising},
pubstate = {published},
tppubtype = {article}
}
The release of the Kinect has fostered the design of novel methods and techniques in several application domains. It has been tested in different contexts, which span from home entertainment to surgical environments. Nonetheless, to promote its adoption to solve real-world problems, the Kinect should be evaluated in terms of precision and accuracy. Up to now, some filtering approaches have been proposed to enhance the precision and accuracy of the Kinect sensor, and preliminary studies have shown promising results. In this work, we discuss the results of a study in which we have compared the most commonly used filtering approaches for Kinect depth data, in both static and dynamic contexts, by using novel metrics. The experimental results show that each approach can be profitably used to enhance the precision and/or accuracy of Kinect depth data in a specific context, whereas the temporal filtering approach is able to reduce noise in different experimental conditions.
2014
Gallo, Luigi
Hand Shape Classification Using Depth Data for Unconstrained 3D Interaction Journal Article
In: Journal of Ambient Intelligence and Smart Environments, vol. 6, no. 1, pp. 93–105, 2014, ISSN: 1876-1364.
Abstract | Links | BibTeX | Tags: 3D interaction, Classification, Kinect, Static hand pose recognition, Touchless interaction, Visualization
@article{galloHandShapeClassification2014,
title = {Hand Shape Classification Using Depth Data for Unconstrained 3D Interaction},
author = { Luigi Gallo},
doi = {10.3233/AIS-130239},
issn = {1876-1364},
year = {2014},
date = {2014-01-01},
journal = {Journal of Ambient Intelligence and Smart Environments},
volume = {6},
number = {1},
pages = {93--105},
abstract = {In this paper, we introduce a novel method for view-independent hand pose recognition from depth data. The proposed approach, which does not rely on color information, provides an estimation of the shape and orientation of the user's hand without constraining him/her to maintain a fixed position in the 3D space. We use principal component analysis to estimate the hand orientation in space, Flusser moment invariants as image features and two SVM-RBF classifiers for visual recognition. Moreover, we describe a novel weighting method that takes advantage of the orientation and velocity of the user's hand to assign a score to each hand shape hypothesis. The complete processing chain is described and evaluated in terms of real-time performance and classification accuracy. As a case study, it has also been integrated into a touchless interface for 3D medical visualization, which allows users to manipulate 3D anatomical parts with up to six degrees of freedom. Furthermore, the paper discusses the results of a user study aimed at assessing if using hand velocity as an indicator of the user's intentionality in changing hand posture results in an overall gain in the classification accuracy. The experimental results show that, especially in the presence of out-of-plane rotations of the hand, the introduction of the velocity-based weighting method produces a significant increase in the pose recognition accuracy.},
keywords = {3D interaction, Classification, Kinect, Static hand pose recognition, Touchless interaction, Visualization},
pubstate = {published},
tppubtype = {article}
}
In this paper, we introduce a novel method for view-independent hand pose recognition from depth data. The proposed approach, which does not rely on color information, provides an estimation of the shape and orientation of the user's hand without constraining him/her to maintain a fixed position in the 3D space. We use principal component analysis to estimate the hand orientation in space, Flusser moment invariants as image features and two SVM-RBF classifiers for visual recognition. Moreover, we describe a novel weighting method that takes advantage of the orientation and velocity of the user's hand to assign a score to each hand shape hypothesis. The complete processing chain is described and evaluated in terms of real-time performance and classification accuracy. As a case study, it has also been integrated into a touchless interface for 3D medical visualization, which allows users to manipulate 3D anatomical parts with up to six degrees of freedom. Furthermore, the paper discusses the results of a user study aimed at assessing if using hand velocity as an indicator of the user's intentionality in changing hand posture results in an overall gain in the classification accuracy. The experimental results show that, especially in the presence of out-of-plane rotations of the hand, the introduction of the velocity-based weighting method produces a significant increase in the pose recognition accuracy.
2012
Essmaeel, Kyis; Gallo, Luigi; Damiani, Ernesto; Pietro, Giuseppe De; Dipand`a, Albert
Multiple Structured Light-Based Depth Sensors for Human Motion Analysis: A Review Proceedings Article
In: Hutchison, David; Kanade, Takeo; Kittler, Josef; Kleinberg, Jon M.; Mattern, Friedemann; Mitchell, John C.; Naor, Moni; Nierstrasz, Oscar; Rangan, C. Pandu; Steffen, Bernhard; Sudan, Madhu; Terzopoulos, Demetri; Tygar, Doug; Vardi, Moshe Y.; Weikum, Gerhard; Bravo, José; Hervás, Ramón; Rodríguez, Marcela (Ed.): Ambient Assisted Living and Home Care, pp. 240–247, Springer Berlin Heidelberg, Vitoria-Gasteiz, Spain, 2012, ISBN: 978-3-642-35394-9 978-3-642-35395-6.
Abstract | Links | BibTeX | Tags: Calibration, Human motion analysis, Interference, Kinect
@inproceedings{essmaeelMultipleStructuredLightBased2012,
title = {Multiple Structured Light-Based Depth Sensors for Human Motion Analysis: A Review},
author = { Kyis Essmaeel and Luigi Gallo and Ernesto Damiani and Giuseppe De Pietro and Albert Dipand{`a}},
editor = { David Hutchison and Takeo Kanade and Josef Kittler and Jon M. Kleinberg and Friedemann Mattern and John C. Mitchell and Moni Naor and Oscar Nierstrasz and C. Pandu Rangan and Bernhard Steffen and Madhu Sudan and Demetri Terzopoulos and Doug Tygar and Moshe Y. Vardi and Gerhard Weikum and José Bravo and Ramón Hervás and Marcela Rodríguez},
doi = {10.1007/978-3-642-35395-6_33},
isbn = {978-3-642-35394-9 978-3-642-35395-6},
year = {2012},
date = {2012-12-01},
urldate = {2016-12-06},
booktitle = {Ambient Assisted Living and Home Care},
volume = {7657},
pages = {240--247},
publisher = {Springer Berlin Heidelberg},
address = {Vitoria-Gasteiz, Spain},
series = {Lecture Notes in Computer Science},
abstract = {Human motion analysis is an increasingly important active research domain with various applications in surveillance, human-machine interaction and human posture analysis. The recent developments in depth sensor technology, especially with the release of the Kinect device, have attracted significant attention to the question of how to take advantage of this technology in order to achieve accurate motion tracking and action detection in marker-less approaches. In this paper, we review the benefits and limitations deriving from the adoption of structured light-based depth sensors in human motion analysis applications. Surveying the relevant literature, we have identified in calibration, interference and bias correction the challenges to tackle for an effective adoption of multi-Kinect systems to improve the visual analysis of human movement.},
keywords = {Calibration, Human motion analysis, Interference, Kinect},
pubstate = {published},
tppubtype = {inproceedings}
}
Human motion analysis is an increasingly important active research domain with various applications in surveillance, human-machine interaction and human posture analysis. The recent developments in depth sensor technology, especially with the release of the Kinect device, have attracted significant attention to the question of how to take advantage of this technology in order to achieve accurate motion tracking and action detection in marker-less approaches. In this paper, we review the benefits and limitations deriving from the adoption of structured light-based depth sensors in human motion analysis applications. Surveying the relevant literature, we have identified in calibration, interference and bias correction the challenges to tackle for an effective adoption of multi-Kinect systems to improve the visual analysis of human movement.
Essmaeel, Kyis; Gallo, Luigi; Damiani, Ernesto; Pietro, Giuseppe De; Dipanda, Albert
Temporal Denoising of Kinect Depth Data Proceedings Article
In: SITIS '12: Eighth International Conference on Signal Image Technology and Internet Based Systems, pp. 47–52, IEEE, Sorrento, Italy, 2012, ISBN: 978-1-4673-5152-2.
Abstract | Links | BibTeX | Tags: Depth instability, Kinect, Smoothing, Temporal denoising
@inproceedings{essmaeelTemporalDenoisingKinect2012,
title = {Temporal Denoising of Kinect Depth Data},
author = { Kyis Essmaeel and Luigi Gallo and Ernesto Damiani and Giuseppe De Pietro and Albert Dipanda},
doi = {10.1109/SITIS.2012.18},
isbn = {978-1-4673-5152-2},
year = {2012},
date = {2012-01-01},
booktitle = {SITIS '12: Eighth International Conference on Signal Image Technology and Internet Based Systems},
pages = {47--52},
publisher = {IEEE},
address = {Sorrento, Italy},
abstract = {The release of the Microsoft Kinect has attracted the attention of researchers in a variety of computer science domains. Even though this device is still relatively new, its recent applications have shown some promising results in terms of replacing current conventional methods like the stereo-camera for robotics navigation, multi-camera system for motion detection and laser scanner for 3D reconstruction. While most work around the Kinect is on how to take full advantage of its capabilities, so far only a few studies have been carried out on the limitations of this device and fewer that provide solutions to enhance the precision of its measurements. In this paper, we review and analyse current work in this area, and present and evaluate a temporal denoising algorithm to reduce the instability of the depth measurements provided by the Kinect over different distances.},
keywords = {Depth instability, Kinect, Smoothing, Temporal denoising},
pubstate = {published},
tppubtype = {inproceedings}
}
The release of the Microsoft Kinect has attracted the attention of researchers in a variety of computer science domains. Even though this device is still relatively new, its recent applications have shown some promising results in terms of replacing current conventional methods like the stereo-camera for robotics navigation, multi-camera system for motion detection and laser scanner for 3D reconstruction. While most work around the Kinect is on how to take full advantage of its capabilities, so far only a few studies have been carried out on the limitations of this device and fewer that provide solutions to enhance the precision of its measurements. In this paper, we review and analyse current work in this area, and present and evaluate a temporal denoising algorithm to reduce the instability of the depth measurements provided by the Kinect over different distances.
2011
Placitelli, Alessio Pierluigi; Gallo, Luigi
3D Point Cloud Sensors for Low-cost Medical In-situ Visualization Proceedings Article
In: 2011 IEEE International Conference on Bioinformatics and Biomedicine Workshops (BIBMW), pp. 596–597, IEEE, Atlanta, GA, USA, 2011, ISBN: 978-1-4577-1613-3.
Abstract | Links | BibTeX | Tags: 3D registration, Augmented Reality, Healthcare, Kinect
@inproceedings{placitelli3DPointCloud2011,
title = {3D Point Cloud Sensors for Low-cost Medical In-situ Visualization},
author = { Alessio Pierluigi Placitelli and Luigi Gallo},
doi = {10.1109/BIBMW.2011.6112435},
isbn = {978-1-4577-1613-3},
year = {2011},
date = {2011-11-01},
booktitle = {2011 IEEE International Conference on Bioinformatics and Biomedicine Workshops (BIBMW)},
pages = {596--597},
publisher = {IEEE},
address = {Atlanta, GA, USA},
abstract = {Medical in-situ visualization deals with the display of the patient's specific imaging data at the location where they actually are. To be effective, it requires high end I/O devices, and computationally expensive and time-consuming calibration and registration steps. In this paper, we explore the use of widely available and low-priced 3D point cloud sensors in medical augmented reality (AR) applications. Specifically, we examine the typical pipeline of AR applications and explore the potential simplifications derived from the use of RGB-D cameras during the calibration and registration steps. Moreover, we describe a low-cost system built from open-source components that takes advantage of 3D point cloud data to apply medical imagery to live-video streams of patients.},
keywords = {3D registration, Augmented Reality, Healthcare, Kinect},
pubstate = {published},
tppubtype = {inproceedings}
}
Medical in-situ visualization deals with the display of the patient's specific imaging data at the location where they actually are. To be effective, it requires high end I/O devices, and computationally expensive and time-consuming calibration and registration steps. In this paper, we explore the use of widely available and low-priced 3D point cloud sensors in medical augmented reality (AR) applications. Specifically, we examine the typical pipeline of AR applications and explore the potential simplifications derived from the use of RGB-D cameras during the calibration and registration steps. Moreover, we describe a low-cost system built from open-source components that takes advantage of 3D point cloud data to apply medical imagery to live-video streams of patients.
Gallo, Luigi; Placitelli, Alessio Pierluigi; Ciampi, Mario
Controller-Free Exploration of Medical Image Data: Experiencing the Kinect Proceedings Article
In: 2011 24th International Symposium on Computer-Based Medical Systems (CBMS), pp. 1–6, IEEE, Bristol, United Kingdom, 2011, ISBN: 978-1-4577-1189-3.
Abstract | Links | BibTeX | Tags: Healthcare, Kinect, Medical Imaging, Touchless interaction
@inproceedings{galloControllerfreeExplorationMedical2011,
title = {Controller-Free Exploration of Medical Image Data: Experiencing the Kinect},
author = { Luigi Gallo and Alessio Pierluigi Placitelli and Mario Ciampi},
doi = {10.1109/CBMS.2011.5999138},
isbn = {978-1-4577-1189-3},
year = {2011},
date = {2011-06-01},
booktitle = {2011 24th International Symposium on Computer-Based Medical Systems (CBMS)},
pages = {1--6},
publisher = {IEEE},
address = {Bristol, United Kingdom},
abstract = {In this paper, an open-source system for a controller-free, highly interactive exploration of medical images is presented. By using a Microsoft Xbox KinectTM as the only input device, the system's user interface allows users to interact at a distance through hand and arm gestures. The paper also details the interaction techniques we have designed specifically for the deviceless exploration of medical imaging data. Since the user interface is touch-free and does not require complex calibration steps, it is suitable for use in operating rooms, where non-sterilizable devices cannot be used.},
keywords = {Healthcare, Kinect, Medical Imaging, Touchless interaction},
pubstate = {published},
tppubtype = {inproceedings}
}
In this paper, an open-source system for a controller-free, highly interactive exploration of medical images is presented. By using a Microsoft Xbox KinectTM as the only input device, the system's user interface allows users to interact at a distance through hand and arm gestures. The paper also details the interaction techniques we have designed specifically for the deviceless exploration of medical imaging data. Since the user interface is touch-free and does not require complex calibration steps, it is suitable for use in operating rooms, where non-sterilizable devices cannot be used.