Computer Graphics and Virtual Reality Lab

Department of Computer Science

Dr Andreas Aristidou

Publications

JOURNALS:

[5]

Andreas Aristidou, Panayiotis Charalambous, Yiorgos Chrysanthou, "Emotion analysis and classification: Understanding the performers emotions using the LMA entities", Submitted for publication in Computer Graphics Forum, September, 2014.
[pdf | -Kb] [Abstract] [BibTeX]

[4]

Andreas Aristidou, Joan Lasenby, Yiorgos Chrysanthou, "Inverse Kinematics techniques in Computer Graphics: a survey", Submitted for publication in IEEE Transactions of Visualisation and Computer Graphics, August, 2014.
[pdf | -Kb] [Abstract] [BibTeX]

[3]

Andreas Aristidou, Yiorgos Chrysanthou, Joan Lasenby, "Extending FABRIK with Model Constraints", Submitted for publication in Computer Animation and Virtual Worlds, February, 2014.
[pdf | -Kb] [Abstract] [BibTeX]

[2]

Andreas Aristidou, Joan Lasenby, "Real-Time Marker Prediction and CoR Estimation in Optical Motion Capture" The Visual Computer, 29(1):7-26, Springer Berlin/Heidelberg, 2013.
[pdf | 1.2Mb] [Abstract] [BibTeX]

This paper addresses the problem of real-time joint localisation of legged skeletons in the presence of missing data. The data is assumed to be labelled 3d marker positions from a motion capture system. An integrated framework is presented which predicts the occluded marker positions using a variable turn model within an Unscented Kalman filter. Inferred information from neighbouring markers is used as observation states; these constraints are efficient, simple and real-time implementable. This work also takes advantage of the common case that missing markers are often visible to only a single camera, resulting in more accurate predictions. An Inverse Kinematics technique is then applied ensuring that the bone lengths remain constant over time; the system can thereby maintain a continuous data-flow. The marker and Centre of Rotation (CoR) positions can be calculated with high accuracy even in cases where markers are occluded for a long period of time. Our methodology is tested against some of the most popular methods for marker prediction and the results confirm that our approach outperforms these methods in estimating both marker and CoR positions.

[1]

Andreas Aristidou, Joan Lasenby, "FABRIK: a fast, iterative solver for the Inverse Kinematics problem", Graphical Models, 73(5):243-260, Elsevier 2011.
[pdf | 1.38Mb] [Abstract] [BibTeX]

Inverse Kinematics is defined as the problem of determining a set of appropriate joint configurations for which the end effectors move to desired positions as smoothly, rapidly, and as accurately as possible. However, many of the currently available methods suffer from high computational cost and production of unrealistic poses. In this paper, a novel heuristic method, called Forward And Backward Reaching Inverse Kinematics (FABRIK), is described and compared with some of the most popular existing methods regarding reliability, computational cost and conversion criteria. FABRIK avoids the use of rotational angles or matrices, and instead finds each joint position via locating a point on a line. Thus, it converges in fewer iterations, has low computational cost and produces visually realistic poses. Constraints can easily be incorporated within FABRIK and multiple chains with multiple end effectors are also easily supported.

 [Back to top]


BOOK CHAPTERS:

[1]

Andreas Aristidou, Joan Lasenby, "Inverse Kinematics solutions using Conformal Geometric Algebra", In L. Dorst and J. Lasenby (Eds), Guide to Geometric Algebra in Practice, Springer Verlag, 2011.
[pdf | 1.09Mb] [ppt | 2.1Mb] [Abstract] [BibTeX]

This paper describes a novel iterative Inverse Kinematics (IK) solver, named FABRIK, that is implemented using Conformal Geometric Algebra (CGA). FABRIK uses a forward and backward iterative approach, finding each joint position via locating a point on a line. We also use the IK of a human hand as an example of implementation where a constrained version of FABRIK was employed for pose tracking. The hand is modelled using CGA, taking advantage of CGA's compact and geometrically intuitive framework and that basic entities in CGA, such as spheres, lines, planes and circles, are simply represented by algebraic objects. This approach can be used in a wide range of computer animation applications and is not limited to the specific problem discussed here. The proposed hand pose tracker is real-time implementable and exploits the advantages of CGA for applications in computer vision, graphics and robotics.

 [Back to top]


CONFERENCE PROCEEDINGS:

[11]

Andreas Aristidou, Efstathios Stavrakis, Yiorgos Chrysanthou, "LMA-Based Motion Retrieval for Folk Dance Cultural Heritage", In Proceedings of the 5th International Conference on Cultural Heritage (EuroMed 2014), Lecture Notes in Computer Science, LNCS volume 8740, pages 207-2016, Limassol, Cyprus, November 3 - 8, 2014.
[pdf | 4.14 Mb] [Abstract] [BibTeX]

Motion capture (mocap) technology is an efficient method for digitizing art-performances, and it is becoming a popular method for the preservation and dissemination of dances. However, stylistic variations of human motion are difficult to measure and cannot be directly extracted from the motion capture data itself. In this work, we present a framework based on Laban Movement Analysis (LMA) that aims to identify style qualities in motion and provides a mechanism for motion indexing using the four LMA components (BODY, EFFORT, SHAPE, SPACE), which can also be subsequently used for intuitive motion retrieval. We have designed and implemented a prototype motion search engine in which users can perform queries using motion clips in a folk dance database. Results demonstrate that the proposed method can be used in place, or in combination with text-based queries, to enable more effective and flexible motion database search and retrieval.

[10]

Andreas Aristidou, Efstathios Stavrakis, Yiorgos Chrysanthou, "Motion Analysis for Folk Dance Evaluation", In Proceedings of the 12th EUROGRAPHICS Workshop on Graphics and Cultural Heritage (GCH), pages 55-64, Darmstadt, Germany, October 5 - 8, 2014.
[pdf | 2.66 Mb] [Abstract] [BibTeX] [BEST PAPER AWARD]

Motion capture techniques are becoming a popular method for digitizing folk dances for preservation and dissemination. Although technically the captured data can be of very high quality, folk dancing, in contrast to choreographed performances, allow for stylistic variations and improvisations that cannot be easily captured by the data themselves. The majority of motion analysis and comparison algorithms are explicitly based on quantitative metrics and thus do not usually provide any insight on style qualities of a performance. In this work, we introduce a motion analysis and comparison framework that is based on Laban Movement Analysis (LMA); these algorithms are particularly useful in the context of teaching folk dances. We present a prototype virtual reality simulator in which users can preview segments of folk dance performed by a 3D avatar and repeat them. The users' performances are captured and subsequently compared to the folk dance template motions. The system then provides intuitive feedback about their performance, which is based on the four LMA components (Body,Effort,Shape,Space) and provides both a quantitative and qualitative evaluation of the performance.

[9]

Andreas Aristidou, Yiorgos Chrysanthou, "Feature extraction for human motion indexing of acted dance performances", In Proceedings of the 9th International Conference on Computer Graphics Theory and Applications (GRAPP'14), pages 277-287, Lisbon, Portugal, January 05-08, 2014.
[pdf | 4.9 Mb] [Abstract] [BibTeX]

There has been an increasing use of pre-recorded motion capture data for animating virtual characters and synthesising different actions; it is although a necessity to establish a resultful method for indexing, classifying and retrieving motion. In this paper, we propose a method that can automatically extract motion qualities from dance performances, in terms of Laban Movement Analysis (LMA), for motion analysis and indexing purposes. The main objectives of this study is to analyse the motion information of different dance performances, using the LMA components, and extract those features that are indicative of certain emotions or actions. LMA encodes motions using four components, Body, Effort, Shape and Space, which represent a wide array of structural, geometric, and dynamic features of human motion. A deeper analysis of how these features change on different movements is presented, investigating the correlations between the performers' acting emotional state and its characteristics, thus indicating the importance and the effect of each feature for the classification of the motion. Understanding the quality of the movement helps to apprehend the intentions of the performer, providing a representative search space for indexing motions.

[8]

Andreas Aristidou, Yiorgos Chrysanthou, "Motion indexing of different emotional states using LMA components", In Proceedings of the ACM SIGGRAPH Asia 2013 conference, Hong Kong, November 19-22, 2013.
[pdf | 1.2Mb] [Abstract] [BibTeX]

Recently, there has been an increasing use of pre-recorded motion capture data, making motion indexing and classification essential for animating virtual characters and synthesising different actions. In this paper, we use a variety of features that encode characteristics of motion using the Body, Effort, Shape and Space components of Laban Movement Analysis (LMA), to explore the motion quality from acted dance performances. Using Principal Component Analysis (PCA), we evaluate the importance of the proposed features - with regards to their ability to separate the performer's emotional state - indicating the weight of each feature in motion classification. PCA has been also used for dimensionality reduction, laying the foundation for the qualitative and quantitative classification of movements based on their LMA characteristics. Early results show that the proposed features provide a representative space for indexing and classification of dance movements with regards to the emotion, which can be used for synthesis and composition purposes.

[7]

Haris Zacharatos, Christos Gatzoulis, Yiorgos Chrysanthou, Andreas Aristidou, "Emotion Recognition for Exergames using Laban Movement Analysis", In Proceedings of ACM Motion in Games 2013 (MIG 2013) conference, Dublin, Ireland, November 7-9, 2013.
[pdf | 320Kb] [Abstract] [BibTeX]

Exergames do not have the capacity to detect whether the players are really enjoying the game-play. The games are not intelligent enough to detect significant emotional states and adapt accordingly in order to offer a better user experience for the players. We propose a set of body motion features, based on the Effort component of Laban Movement Analysis (LMA), that are used to provide sets of classifiers for emotion recognition in a game scenario for four emotional states: concentration, meditation, excitement and frustration. Experimental results show that, the system is capable of successfully recognizing the four different emotional states at a very high rate.

[6]

Efstathios Stavrakis, Andreas Aristidou, Maria Savva, Stephania Loizidou Himona, Yiorgos Chrysanthou, "Digitization of Cypriot Folk Dances", In Proceedings of the 4th International Conference in Cultural Heritage Preservation (EuroMed 2012), Lecture Notes in Computer Science, LNCS volume 7616, pages 404-413, Limassol, Cyprus, October 29 - November 3, 2012.
[pdf | 618Kb] [Abstract] [BibTeX]

In this article an initiative to preserve the Cypriot folk dance heritage is reported. The project aims to create a publicly accessible digital archive of folk dances that does not only include video recordings, commonly used to document dance performances. In addition to rare video material held by local cultural institutions, state-of-the-art motion capture technologies are utilized to record and archive high quality motion data of expert dancers performing these traditional dances. Apart from the goal of preserving this intangible cultural heritage by digitizing it, the project is interested in increasing the awareness of the local community to its dance heritage. To achieve this a 3D video game for children is developed to teach these folk dances to the younger generations.

[5]

Andreas Aristidou, Joan Lasenby, "Motion Capture with Constrained Inverse Kinematics for Real-Time Hand Tracking", In IEEE Proceedings of the 4th International Symposium on Communications, Control and Signal Processing (ISCCSP'10), Limassol, Cyprus, May 3-5, 2010.
[pdf | 354Kb] [Abstract] [BibTeX]

Articulated hand tracking systems have been commonly used in virtual reality applications, including systems with human-computer interaction or interaction with game consoles. However, building an effective real-time hand pose tracker remains challenging. In this , we present a simple and efficient methodology for tracking and reconstructing 3d hand poses using a markered optical motion capture system. Markers were positioned at strategic points, and an inverse kinematics solver was incorporated to fit the rest of the joints to the hand model. The model is highly constrained with rotational and orientational constraints, allowing motion only within a feasible set. The method is real-time implementable and the results are promising, even with a low frame rate.

[4]

Andreas Aristidou, Jonathan Cameron, Joan Lasenby, "Predicting Missing Markers to Drive Real-Time Centre of Rotation Estimation", In Proceedings of the 5th International Conference on Articulated Motion and Deformable Objects, AMDO'08, Lecture Notes in Computer Science, LNCS volume 5098, pages 239-247, Mallorca, Spain, July 9-11, 2008.
[pdf | 1Mb] [Abstract] [BibTeX]

This paper addresses the problem of real-time location of the joints or centres of rotation (CoR) of human skeletons in the presence of missing data. The data is assumed to be 3d marker positions from a motion capture system. We present an integrated framework which predicts the occluded marker positions using a Kalman filter in combination with inferred information from neighbouring markers and thereby maintains a continuous data-flow. The CoR positions can be calculated with high accuracy even in cases where markers are occluded for a long period of time.

[3]

Andreas Aristidou, Jonathan Cameron, Joan Lasenby, "Real-Time Estimation of Missing Markers in Human Motion Capture", In IEEE Proceedings of the 2nd International Conference on Bioinformatics and Biomedical Engineering, iCBBE'08, pages 1343-1346, Shanghai, China, May 16-18, 2008.
[pdf | 472Kb] [Abstract] [BibTeX]

This paper considers the problem of taking marker locations from optical motion capture data to identify and parameterise the underlying human skeleton structure and motion over time. It is concerned with real-time algorithms suitable for use within a visual feedback system. A common problem in motion capture is marker occlusion. Most current methods are only useful for offline processing or become ineffective when a significant portion of markers are missing for a long period of time. This paper presents a prediction algorithm, using a Kalman filter approach in combination with inferred information from neighbouring markers, to provide a continuous flow of data. The results are accurate and reliable even in cases where all markers on a limb are occluded, or one or two markers are not visible for a large sequence of frames. Pre-defined models are not required and skeleton fitting to this complete data can then be updated in real-time.

[2]

Andreas Aristidou, Paul Pangalos, Hamid Aghvami, "Tracking Multiple Sports Players for Mobile Display", In Proceeding of the 2007 International Conference on Image Processing, Computer Vision, and Pattern Recognition, IPCV'07, pages 53-59, Las Vegas, Nevada, USA, June 25-28, 2007.
[pdf | 353Kb] [Abstract] [BibTeX]

An architecture system and a method for tracking people are presented for sports applications. The system's input is video data from static camera and the output is the real world, real-time positions of the players during a sport event. This output can be used for low-bandwidth match play animations for web or wireless display, and for analysis of fitness and tactics of the teams and players. Firstly, an efficient real-time background modeling and maintenance is described based on the segmentation of input images into stationary and non-stationary blocks. In order to detect foreground objects a method based on background subtraction is implemented, using chromaticity and lightness, with the intention of avoiding shadows and jitters. Object tracking is achieved using a cost function of blob information such as the centroid coordinates, the covered area, the velocity and the color. A condensation filter is also implemented in order to cope with complex cases such as when regions enter/exit the scene and when they can be occluded by other regions. The results of the proposed methods, the players' location and trajectories are presented and discussed.

[1]

Theodoros Giannakopoulos, Dimitrios Kosmopoulos, Andreas Aristidou, Sergios Theodoridis, "Violence Content Classification using Audio Features", Hellenic Artificial Intelligence Conference SETN-06, Lecture Notes in Computer Science (LNAI), Volume 3955, pages 502-507, Heraklion, Crete, Greece, May 18-20, 2006.
[pdf | 231Kb] [Abstract] [BibTeX]

This work studies the problem of violence detection in audio data, which can be used for automated content rating. We employ some popular frame-level audio features both from the time and frequency domain. Afterwards, several statistics of the calculated feature sequences are fed as input to a Support Vector Machine classifier, which decides about the segment content with respect to violence. The presented experimental results verify the validity of the approach and exhibit a better performance than the other known approaches.

 [Back to top]


OTHER PUBLICATIONS:

[2]

Andreas Aristidou, Efstathios Stavrakis, Yiorgos Chrysanthou, "Cypriot Intangible Cultural Heritage: Digitizing Folk Dances", Cyprus Computer Society journal <<Πληροφορική>>, Issue 25, pages 42-49, Nicosia, Cyprus, April 2014.
[pdf | 230Kb] [pdf | 325Kb] [Abstract] [BibTeX]

Cyprus has a long and rich history of dance tradition which unfortunately, year after year, tends to be forgotten; thus, it is our duty to help documenting and disseminating our dance heritage to the younger generations. In this work, we aim to preserve the Cypriot folk dance heritage, creating a state-of-the-art publicly accessible digital archive of folk dances. Our dance library, apart from the rare video materials that are commonly used to document dance performances, utilises three dimensional motion capture technologies to record and archive high quality motion data of expert dancers.

[1]

Andreas Aristidou, "Reliable indexing and recognition of violent scenes using audio features", Annual journal of the Department of Informatics and Telecommunications, Best dissertations of the year, Athens, Greece, 2006.
[pdf in Greek | 330Kb] [Abstract] [BibTeX]

Η εργασία αυτή επικεντρώνεται στην μελέτη των μεθόδων που αποσκοπούν στην αυτόματη και αξιόπιστη αναγνώριση των σκηνών βίας με χρήση της ακουστικής πληροφορίας του σήματος. Χρησιμοποιήθηκαν συγκεκριμένα δημοφιλή ηχητικά χαρακτηριστικά όπως η ενεργειακή εντροπία, η ενέργεια μικρής διάρκειας (short time energy), το ZCR, η φασματική διακύμανση (spectral flux). Στη συνέχεια, αναπτύξαμε ένα αποτελεσματικό μηχανισμό ταξινόμησης SVM (Support Vector Machine), ο οποίος εκπαιδεύθηκε αξιοποιώντας αυτά τα ηχητικά χαρακτηριστικά γνωρίσματα της βίας. Τα αποτελέσματα αυτής της ταξινόμησης καταγράφονται, αξιολογούνται οι αλγόριθμοι και αναφέρονται η απόδοση και το ποσοστό επιτυχούς ταξινόμησης, το οποίο είναι καλύτερο από οποιαδήποτε άλλη προτεινόμενη μέθοδο.

 [Back to top]


REPORTS:

[6]

Andreas Aristidou, Yiorgos Chrysanthou, "Marker Prediction and Skeletal Reconstruction in Motion Capture Technology", A technical report (UCY-CS-TR-13-2) submitted to the Department of Computer Science at University of Cyprus, Nicosia, Cyprus, August, 2013.
[pdf | 1.83Mb] [Abstract] [BibTeX]

Optical motion capture systems suffer from marker occlusions resulting in loss of useful information. This technical report addresses the problem of real-time joint localisation of human skeletons in the presence of such missing data; at least three markers are placed at strategic positions on each limb segment. The data is assumed to be labelled 3d marker positions from an 8-cameras PhaseSpace Impulse X2 motion capture system. An integrated framework is implemented using MATLAB which predicts the occluded marker positions using a Variable Turn Model within an Unscented Kalman filter. Inferred information from neighbouring markers is used as observation states; these constraints are efficient, simple, and real-time implementable. This work also takes advantage of the common case that missing markers are still visible to a single camera, by combining predictions with under-determined positions, resulting in more accurate predictions. An Inverse Kinematics (IK) technique is then applied ensuring that the bone lengths remain constant over time; the human skeleton has been structured hierarchically and the IK solver has been applied sequentially to the kinematic chains, maintaining a continuous data-flow. The marker and Centre of Rotation (CoR) positions can be calculated with high accuracy even in cases where markers are occluded for a long period of time. Results demonstrate the efficiency of both the proposed methodology and the implemented algorithms in MATLAB.

Andreas Aristidou, "Tracking and Modelling Motion for Biomechanical Analysis", A dissertation submitted to the University of Cambridge for the Degree of Doctor of Philosophy, Cambridge, October 2010.
Supervisor: Dr Joan Lasenby, Examiners: Prof William J. Fitzgerald & Prof Adrian Hilton.
[pdf | 23.9Mb] [Abstract] [BibTeX]

This thesis focuses on the problem of determining appropriate skeletal configurations for which a virtual animated character moves to desired positions as smoothly, rapidly, and as accurately as possible. During the last decades, several methods and techniques, sophisticated or heuristic, have been presented to produce smooth and natural solutions to the Inverse Kinematics (IK) problem. However, many of the currently available methods suffer from high computational cost and production of unrealistic poses. In this study, a novel heuristic method, called Forward And Backward Reaching Inverse Kinematics (FABRIK), is proposed, which returns visually natural poses in real-time, equally comparable with highly sophisticated approaches. It is capable of supporting constraints for most of the known joint types and it can be extended to solve problems with multiple end effectors, multiple targets and closed loops. FABRIK was compared against the most popular IK approaches and evaluated in terms of its robustness and performance limitations. This thesis also includes a robust methodology for marker prediction under multiple marker occlusion for extended time periods, in order to drive real-time centre of rotation (CoR) estimations. Inferred information from neighbouring markers has been utilised, assuming that the inter-marker distances remain constant over time. This is the first time where the useful information about the missing markers positions which are partially visible to a single camera is deployed. Experiments demonstrate that the proposed methodology can effectively track the occluded markers with high accuracy, even if the occlusion persists for extended periods of time, recovering in real-time good estimates of the true joint positions. In addition, the predicted positions of the joints were further improved by employing FABRIK to relocate their positions and ensure a fixed bone length over time. Our methodology is tested against some of the most popular methods for marker prediction and the results confirm that our approach outperforms these methods in estimating both marker and CoR positions. Finally, an efficient model for real-time hand tracking and reconstruction that requires a minimum number of available markers, one on each finger, is presented. The proposed hand model is highly constrained with joint rotational and orientational constraints, restricting the fingers and palm movements to an appropriate feasible set. FABRIK is then incorporated to estimate the remaining joint positions and to fit them to the hand model. Physiological constraints, such as inertia, abduction, flexion etc, are also incorporated to correct the final hand posture. A mesh deformation algorithm is then applied to visualise the movements of the underlying hand skeleton for comparison with the true hand poses. The mathematical framework used for describing and implementing the techniques discussed within this thesis is Conformal Geometric Algebra (CGA).

[4]

Andreas Aristidou, Joan Lasenby, "Inverse Kinematics: a review of existing techniques and introduction of a new fast iterative solver", A technical report (CUEDF-INFENG, TR-632) submitted to the Department of Information Engineering at University of Cambridge, Cambridge, September 2009.
[pdf | 7.9Mb] [Abstract] [BibTeX]

Inverse Kinematics (IK) is defined as the problem of determining a set of appropriate joint configurations for which the end effectors move to desired positions as smoothly, rapidly, and as accurately as possible. During the last decades, several methods and techniques, sophisticated or heuristic, have been presented to produce fast and realistic solutions to the IK problem. However, most of the currently available methods suffer from high computational cost and production of unrealistic poses. This report reviews and compares the most popular IK methods regarding reliability, computational cost and conversion criteria, with a new heuristic iterative method, called Forward And Backward Reaching Inverse Kinematics (FABRIK). FABRIK avoids the use of rotational angles or matrices, and instead finds each joint position via locating a point on a line. Thus, it converges in fewer iterations, has low computational cost and produces realistic poses. Constraints can easily be incorporated within the FABRIK methodology and multiple chains with multiple end effectors are also easily supported.

[3]

Andreas Aristidou, Joan Lasenby, Jonathan Cameron, "Methods for Real-time Restoration and Estimation in Optical Motion Capture", A technical report (CUEDF-INFENG, TR-619) submitted to the Department of Information Engineering at University of Cambridge, Cambridge, January 2009.
[pdf | 1.81Mb] [Abstract] [BibTeX]

This report considers the problem of using marker locations from optical motion capture data to identify and parameterise the underlying human skeleton structure and motion over time. It is concerned with real-time algorithms suitable for use within a visual feedback system. The algorithms presented require 3 markers on each limb segment and can be implemented in a sequential fashion; hence the computational cost of updating the centres of rotation is independent of the number of data points previously available. However, a common problem in motion capture is marker occlusion. Most current methods are only useful for offline processing or become ineffective when a significant proportion of markers are missing for a long period of time. This report presents an integrated framework which predicts the occluded marker positions using a Kalman filter in combination with inferred information from neighbouring markers and thereby maintains a continuous data-flow. The results are accurate and reliable even in cases where all markers on a limb segment are occluded, or one or two markers are non-visible for a large sequence of frames. Moreover, it investigates several extensions of the proposed methodology for real-time applications using a sophisticated variable turn model, together with rotor prediction and unscented theory for filtering. It compares each individual method according to its accuracy, complexity and processing time and suggests the optimal solution for each case. Pre-defined models are not required and skeleton fitting to this complete data can then be updated in real-time.

[2]

Andreas Aristidou, "Real-Time Estimation of Missing Data and Localisation of the Centre of Rotation in Optical Motion Capture Data", A report submitted to the University of Cyprus for the first year of the degree of Doctor of Philosophy, Cambridge, March 2008.
[pdf | 1.41Mb] [Abstract]

This project considers the problem of taking marker locations from optical motion capture data to identify and parameterise the underlying human skeleton structure and motion over time. It is concerned with real-time algorithms suitable for use within a visual feedback system. The algorithms presented require 3 markers on each limb segment and can be implemented in a sequential fashion; hence the computational cost of updating the centres of rotation is independent of the number of data points previously available. However, a common problem in motion capture is marker occlusion. Most current methods are only useful for offline processing or become ineffective when a significant proportion of markers are missing for a long period of time. This paper presents an integrated framework which predicts the occluded marker positions using a Kalman filter in combination with inferred information from neighbouring markers and thereby maintains a continues data-flow. The results are accurate and reliable even in cases where all markers on a limb segment are occluded, or one or two markers are non-visible for a large sequence of frames. Pre-defined models are not required and skeleton fitting to this complete data can then be updated in real-time.

[1]

Andreas Aristidou, "A robust method for tracking sports players for web or wireless display", A dissertation submitted to the University of London, King's College London, for the Degree of Master of Science, London, September 2006.
[pdf | 1.15Mb] [Abstract]

An architecture system and a method for tracking people are presented for sports applications. The system's input is video data from static camera of a stadium and the output is the real world, real-time positions of the players during a sport event. This output can be used for low-bandwidth match play animations for web or wireless display, and for analysis of fitness and tactics of the teams and players. Firstly, in this study, a real-time background modelling and maintenance is described based on the segmentation of input images into stationary and non-stationary blocks. This method is efficient even if many objects are simultaneously visible and the background can be seen only for a short time of period. In order to detect foreground objects a method based on background subtraction is implemented, using chromaticity and lightness, with the intention of avoiding shadows and jitters. Furthermore, object tracking is achieved using a cost function of blob information such as the centroid coordinates, the covered area, the velocity and the colour. In addition to this, a condensation filter is implemented in order to cope with complex cases such as when regions enter/exit the scene and when they can be occluded by other regions. Thereafter, a field estimation algorithm, which utilizes the colour information of the ground is proposed and a new robust projective transformation from real image coordinates to the upper-view, virtual ground image is applied. Finally, the results of the proposed methods are presented individually in addition to the trajectories of moving objects in web/wireless applications such as a Java Applet.

 [Back to top]