http://hdl.handle.net/2117/2251 Wed, 22 May 2013 04:33:42 GMT 2013-05-22T04:33:42Z webmaster.bupc@upc.edu Universitat Politècnica de Catalunya. Servei de Biblioteques i Documentació no http://hdl.handle.net/2117/15545 Title: Aforismo 74: no te pongas intratable con quienes fuiste agradable Authors: Torras, Carme Abstract: Las necedades siempre sorprenden a todos, pues el necio es audaz en atrevimiento. Su torpeza le impide advertir que desentonará con su conducta, y eso mismo le quita la vergüenza de hacer el ridículo. En cambio, el hombre de cordura entra con gran cuidado. Su escudo es la advertencia y el recato, y va observando y descubriendo lo que hay en el ambiente, para actuar con el mínimo de riesgo. Todo atrevimiento que carezca de reflexión está condenado al despeño, aunque tal vez lo salve el azar venturoso. Conviene nadar con cuidado en aguas que se temen hondas: ve probando poco a poco con sagacidad y ganando terreno con prudencia. Hay grandes confusiones hoy en el trato humano: conviene ir siempre tirando sondas que vayan orientándote. Mon, 12 Mar 2012 18:22:54 GMT http://hdl.handle.net/2117/15545 2012-03-12T18:22:54Z Torras, Carme no learning (artificial intelligence) Las necedades siempre sorprenden a todos, pues el necio es audaz en atrevimiento. Su torpeza le impide advertir que desentonará con su conducta, y eso mismo le quita la vergüenza de hacer el ridículo. En cambio, el hombre de cordura entra con gran cuidado. Su escudo es la advertencia y el recato, y va observando y descubriendo lo que hay en el ambiente, para actuar con el mínimo de riesgo. Todo atrevimiento que carezca de reflexión está condenado al despeño, aunque tal vez lo salve el azar venturoso. Conviene nadar con cuidado en aguas que se temen hondas: ve probando poco a poco con sagacidad y ganando terreno con prudencia. Hay grandes confusiones hoy en el trato humano: conviene ir siempre tirando sondas que vayan orientándote. http://hdl.handle.net/2117/9181 Title: Local boosted features for pedestrian detection Authors: Villamizar Vergel, Michael Alejandro; Sanfeliu Cortés, Alberto; Andrade-Cetto, Juan Abstract: The present paper addresses pedestrian detection using local boosted features that are learned from a small set of training images. Our contribution is to use two boosting steps. The first one learns discriminant local features corresponding to pedestrian parts and the second one selects and combines these boosted features into a robust class classifier. In contrast of other works, our features are based on local differences over Histograms of Oriented Gradients (HoGs). Experiments carried out to a public dataset of pedestrian images show good performance with high classification rates Wed, 29 Sep 2010 17:15:00 GMT http://hdl.handle.net/2117/9181 2010-09-29T17:15:00Z Villamizar Vergel, Michael Alejandro; Sanfeliu Cortés, Alberto; Andrade-Cetto, Juan no The present paper addresses pedestrian detection using local boosted features that are learned from a small set of training images. Our contribution is to use two boosting steps. The first one learns discriminant local features corresponding to pedestrian parts and the second one selects and combines these boosted features into a robust class classifier. In contrast of other works, our features are based on local differences over Histograms of Oriented Gradients (HoGs). Experiments carried out to a public dataset of pedestrian images show good performance with high classification rates http://hdl.handle.net/2117/7811 Title: A recursive embedding approach to median graph computation Authors: Ferrer Sumsi, Miquel; Valveny, Ernest; Bunke, Horst Abstract: The median graph has been shown to be a good choice to infer a representative of a set of graphs. It has been successfully applied to graph-based classification and clustering. Nevertheless, its computation is extremely complex. Several approaches have been presented up to now based on different strategies. In this paper we present a new approximate recursive algorithm for median graph computation based on graph embedding into vector spaces. Preliminary experiments on three databases show that this new approach is able to obtain better medians than the previous existing approaches. Wed, 23 Jun 2010 10:24:44 GMT http://hdl.handle.net/2117/7811 2010-06-23T10:24:44Z Ferrer Sumsi, Miquel; Valveny, Ernest; Bunke, Horst no The median graph has been shown to be a good choice to infer a representative of a set of graphs. It has been successfully applied to graph-based classification and clustering. Nevertheless, its computation is extremely complex. Several approaches have been presented up to now based on different strategies. In this paper we present a new approximate recursive algorithm for median graph computation based on graph embedding into vector spaces. Preliminary experiments on three databases show that this new approach is able to obtain better medians than the previous existing approaches. http://hdl.handle.net/2117/7485 Title: Graph-based k-means clustering: A comparison of the set versus the generalized median graph Authors: Ferrer Sumsi, Miquel; Valveny, Ernest; Serratosa Casanelles, Francesc; Bardaji Goikoetxea, Itziar; Bunke, Horst Abstract: In this paper we propose the application of the generalized median graph in a graph-based k-means clustering algorithm. In the graph-based k-means algorithm, the centers of the clusters have been traditionally represented using the set median graph. We propose an approximate method for the generalized median graph computation that allows to use it to represent the centers of the clusters. Experiments on three databases show that using the generalized median graph as the clusters representative yields better results than the set median graph. Wed, 02 Jun 2010 12:49:21 GMT http://hdl.handle.net/2117/7485 2010-06-02T12:49:21Z Ferrer Sumsi, Miquel; Valveny, Ernest; Serratosa Casanelles, Francesc; Bardaji Goikoetxea, Itziar; Bunke, Horst no In this paper we propose the application of the generalized median graph in a graph-based k-means clustering algorithm. In the graph-based k-means algorithm, the centers of the clusters have been traditionally represented using the set median graph. We propose an approximate method for the generalized median graph computation that allows to use it to represent the centers of the clusters. Experiments on three databases show that using the generalized median graph as the clusters representative yields better results than the set median graph. http://hdl.handle.net/2117/7110 Title: 3D object reconstruction from Swissranger sensor data using a spring-mass model Authors: Dellen, Babette; Alenyà Ribas, Guillem; Foix Salmerón, Sergi; Torras, Carme Abstract: We register close-range depth images of objects using a Swissranger sensor and apply a spring-mass model for 3D object reconstruction. The Swissranger sensor delivers depth images in real time which have, compared with other types of sensors, such as laser scanners, a lower resolution and are afflicted with larger uncertainties. To reduce noise and remove outliers in the data, we treat the point cloud as a system of interacting masses connected via elastic forces. We investigate two models, one with and one without a surface-topology preserving interaction strength. The algorithm is applied to synthetic and real Swissranger sensor data, demonstrating the feasibility of the approach. This method represents a preliminary step before fitting higher-level surface descriptors to the data, which will be required to define object-action complexes (OACS) for robot applications. Mon, 03 May 2010 15:54:40 GMT http://hdl.handle.net/2117/7110 2010-05-03T15:54:40Z Dellen, Babette; Alenyà Ribas, Guillem; Foix Salmerón, Sergi; Torras, Carme no Swissranger sensor 3D reconstruction Spring-mass model We register close-range depth images of objects using a Swissranger sensor and apply a spring-mass model for 3D object reconstruction. The Swissranger sensor delivers depth images in real time which have, compared with other types of sensors, such as laser scanners, a lower resolution and are afflicted with larger uncertainties. To reduce noise and remove outliers in the data, we treat the point cloud as a system of interacting masses connected via elastic forces. We investigate two models, one with and one without a surface-topology preserving interaction strength. The algorithm is applied to synthetic and real Swissranger sensor data, demonstrating the feasibility of the approach. This method represents a preliminary step before fitting higher-level surface descriptors to the data, which will be required to define object-action complexes (OACS) for robot applications. http://hdl.handle.net/2117/7108 Title: Simulating dynamical systems for early vision Authors: Dellen, Babette; Wörgötter, Florentin Abstract: We propose a novel algorithm for stereo matching using a dynamical systems approach. The stereo correspondence problem is first formulated as an energy minimization problem. From the energy function, we derive a system of differential equations describing the corresponding dynamical system of interacting elements, which we solve using numerical integration. Optimization is introduced by means of a damping term and a noise term, an idea similar to simulated annealing. The algorithm is tested on the Middlebury stereo benchmark. Mon, 03 May 2010 15:27:24 GMT http://hdl.handle.net/2117/7108 2010-05-03T15:27:24Z Dellen, Babette; Wörgötter, Florentin no We propose a novel algorithm for stereo matching using a dynamical systems approach. The stereo correspondence problem is first formulated as an energy minimization problem. From the energy function, we derive a system of differential equations describing the corresponding dynamical system of interacting elements, which we solve using numerical integration. Optimization is introduced by means of a damping term and a noise term, an idea similar to simulated annealing. The algorithm is tested on the Middlebury stereo benchmark. http://hdl.handle.net/2117/7107 Title: Median graph computation by means of a genetic approach based on minimum common supergraph and maximum common subraph Authors: Ferrer Sumsi, Miquel; Valveny, Ernest; Serratosa Casanelles, Francesc Abstract: Given a set of graphs, the median graph has been theoretically presented as a useful concept to infer a representative of the set. However, the computation of the median graph is a highly complex task and its practical application has been very limited up to now. In this work we present a new genetic algorithm for the median graph computation. A set of experiments on real data, where none of the existing algorithms for the median graph computation could be applied up to now due to their computational complexity, show that we obtain good approximations of the median graph. Finally, we use the median graph in a real nearest neighbour classification showing that it leaves the box of the only-theoretical concepts and demonstrating, from a practical point of view, that can be a useful tool to represent a set of graphs. Mon, 03 May 2010 13:01:33 GMT http://hdl.handle.net/2117/7107 2010-05-03T13:01:33Z Ferrer Sumsi, Miquel; Valveny, Ernest; Serratosa Casanelles, Francesc no Pattern recognition Given a set of graphs, the median graph has been theoretically presented as a useful concept to infer a representative of the set. However, the computation of the median graph is a highly complex task and its practical application has been very limited up to now. In this work we present a new genetic algorithm for the median graph computation. A set of experiments on real data, where none of the existing algorithms for the median graph computation could be applied up to now due to their computational complexity, show that we obtain good approximations of the median graph. Finally, we use the median graph in a real nearest neighbour classification showing that it leaves the box of the only-theoretical concepts and demonstrating, from a practical point of view, that can be a useful tool to represent a set of graphs. http://hdl.handle.net/2117/2689 Title: Improving background subtraction based on a casuistry of colour-motion segmentation problems Authors: Huerta Casado, Iván; Rowe, Daniel; Mozerov, Mikhail; Gonzàlez, Jordi Abstract: The basis for the high-level interpretation of observed patterns of human motion still relies on motion segmentation. Popular approaches based on background subtraction use colour information to model each pixel during a training period. Nevertheless, a deep analysis on colour segmentation problems demonstrates that colour segmentation is not enough to detect all foreground objects in the image, for instance when there is a lack of colour necessary to build the background model. In this paper, our segmentation procedure is based not only on colour, but also on intensity information. Consequently, the intensity model enhances segmentation when the use of colour is not feasible. Experimental results demonstrate the feasibility of our approach. Fri, 13 Mar 2009 09:22:02 GMT http://hdl.handle.net/2117/2689 2009-03-13T09:22:02Z Huerta Casado, Iván; Rowe, Daniel; Mozerov, Mikhail; Gonzàlez, Jordi no The basis for the high-level interpretation of observed patterns of human motion still relies on motion segmentation. Popular approaches based on background subtraction use colour information to model each pixel during a training period. Nevertheless, a deep analysis on colour segmentation problems demonstrates that colour segmentation is not enough to detect all foreground objects in the image, for instance when there is a lack of colour necessary to build the background model. In this paper, our segmentation procedure is based not only on colour, but also on intensity information. Consequently, the intensity model enhances segmentation when the use of colour is not feasible. Experimental results demonstrate the feasibility of our approach. http://hdl.handle.net/2117/2688 Title: Hierarchical eyelid and face tracking Authors: Orozco, Francisco J.; Gonzàlez, Jordi; Rius, Ignasi; Roca, Francesc Xavier Abstract: Most applications on Human Computer Interaction (HCI) require to extract the movements of user faces, while avoiding high memory and time expenses. Moreover, HCI systems usually use low-cost cameras, while current face tracking techniques strongly depend on the image resolution. In this paper, we tackle the problem of eyelid tracking by using Appearance-Based Models, thus achieving accurate estimations of the movements of the eyelids, while avoiding cues, which require high-resolution faces, such as edge detectors or colour information. Consequently, we can track the fast and spontaneous movements of the eyelids, a very hard task due to the small resolution of the eye regions. Subsequently, we combine the results of eyelid tracking with the estimations of other facial features, such as the eyebrows and the lips. As a result, a hierarchical tracking framework is obtained: we demonstrate that combining two appearance-based trackers allows to get accurate estimates for the eyelid, eyebrows, lips and also the 3D head pose by using low-cost video cameras and in real-time. Therefore, our approach is shown suitable to be used for further facial-expression analysis. Fri, 13 Mar 2009 09:21:53 GMT http://hdl.handle.net/2117/2688 2009-03-13T09:21:53Z Orozco, Francisco J.; Gonzàlez, Jordi; Rius, Ignasi; Roca, Francesc Xavier no Most applications on Human Computer Interaction (HCI) require to extract the movements of user faces, while avoiding high memory and time expenses. Moreover, HCI systems usually use low-cost cameras, while current face tracking techniques strongly depend on the image resolution. In this paper, we tackle the problem of eyelid tracking by using Appearance-Based Models, thus achieving accurate estimations of the movements of the eyelids, while avoiding cues, which require high-resolution faces, such as edge detectors or colour information. Consequently, we can track the fast and spontaneous movements of the eyelids, a very hard task due to the small resolution of the eye regions. Subsequently, we combine the results of eyelid tracking with the estimations of other facial features, such as the eyebrows and the lips. As a result, a hierarchical tracking framework is obtained: we demonstrate that combining two appearance-based trackers allows to get accurate estimates for the eyelid, eyebrows, lips and also the 3D head pose by using low-cost video cameras and in real-time. Therefore, our approach is shown suitable to be used for further facial-expression analysis. http://hdl.handle.net/2117/2687 Title: Robust multiple-people tracking using color-based particle filters Authors: Rowe, Daniel; Huerta Casado, Iván; Gonzàlez, Jordi; Villanueva, Juan J. Abstract: Robust and accurate people tracking is a key task in many promising computer-vision applications. One must deal with non-rigid targets in open-world scenarios, whose shape and appearance evolve over time. Targets may interact, causing partial or complete occlusions. This paper improves tracking by means of particle filtering, where occlusions are handled considering the target's predicted trajectories. Model drift is tackled by careful updating, based on the history of likelihood measures. A colour-based likelihood, computed from histogram similarity, is used. Experiments are carried out using sequences from the CAVIAR database. Fri, 13 Mar 2009 09:21:44 GMT http://hdl.handle.net/2117/2687 2009-03-13T09:21:44Z Rowe, Daniel; Huerta Casado, Iván; Gonzàlez, Jordi; Villanueva, Juan J. no Robust and accurate people tracking is a key task in many promising computer-vision applications. One must deal with non-rigid targets in open-world scenarios, whose shape and appearance evolve over time. Targets may interact, causing partial or complete occlusions. This paper improves tracking by means of particle filtering, where occlusions are handled considering the target's predicted trajectories. Model drift is tackled by careful updating, based on the history of likelihood measures. A colour-based likelihood, computed from histogram similarity, is used. Experiments are carried out using sequences from the CAVIAR database. http://hdl.handle.net/2117/2686 Title: Natural language descriptions of human behaviour from video sequences Authors: Fernández, Carles; Baiget, Pau; Roca, Francesc Xavier; Gonzàlez, Jordi Abstract: This contribution addresses the generation of textual descriptions in several natural languages for evaluation of human behavior in video sequences. The problem is tackled by converting geometrical information extracted from videos of the scenario into predicates in fuzzy logic formalism, which facilitates the internal representations of the conceptual data and allows the temporal analysis of situations in a deterministic fashion, by means of Situation Graph Trees (SGTs). The results of the analysis are stored in structures proposed by the Discourse Representation Theory (DRT), which facilitate a subsequent generation of natural language text. This set of tools has been proved to be perfectly suitable for the specified purpose. Fri, 13 Mar 2009 09:21:35 GMT http://hdl.handle.net/2117/2686 2009-03-13T09:21:35Z Fernández, Carles; Baiget, Pau; Roca, Francesc Xavier; Gonzàlez, Jordi no This contribution addresses the generation of textual descriptions in several natural languages for evaluation of human behavior in video sequences. The problem is tackled by converting geometrical information extracted from videos of the scenario into predicates in fuzzy logic formalism, which facilitates the internal representations of the conceptual data and allows the temporal analysis of situations in a deterministic fashion, by means of Situation Graph Trees (SGTs). The results of the analysis are stored in structures proposed by the Discourse Representation Theory (DRT), which facilitate a subsequent generation of natural language text. This set of tools has been proved to be perfectly suitable for the specified purpose. http://hdl.handle.net/2117/2685 Title: Semantic annotation of complex human scenes for multimedia surveillance Authors: Fernández, Carles; Baiget, Pau; Roca, Francesc Xavier; Gonzàlez, Jordi Abstract: A Multimedia Surveillance System (MSS) is considered for automatically retrieving semantic content from complex outdoor scenes, involving both human behavior and traffic domains. To characterize the dynamic information attached to detected objects, we consider a deterministic modeling of spatio-temporal features based on abstraction processes towards fuzzy logic formalism. A situational analysis over conceptualized information will not only allow us to describe human actions within a scene, but also to suggest possible interpretations of the behaviors perceived, such as situations involving thefts or dangers of running over. Towards this end, the different levels of semantic knowledge implied throughout the process are also classified into a proposed taxonomy. Fri, 13 Mar 2009 09:21:27 GMT http://hdl.handle.net/2117/2685 2009-03-13T09:21:27Z Fernández, Carles; Baiget, Pau; Roca, Francesc Xavier; Gonzàlez, Jordi no A Multimedia Surveillance System (MSS) is considered for automatically retrieving semantic content from complex outdoor scenes, involving both human behavior and traffic domains. To characterize the dynamic information attached to detected objects, we consider a deterministic modeling of spatio-temporal features based on abstraction processes towards fuzzy logic formalism. A situational analysis over conceptualized information will not only allow us to describe human actions within a scene, but also to suggest possible interpretations of the behaviors perceived, such as situations involving thefts or dangers of running over. Towards this end, the different levels of semantic knowledge implied throughout the process are also classified into a proposed taxonomy. http://hdl.handle.net/2117/2684 Title: Ontology for semantic integration in a cognitive surveillance system Authors: Fernández, Carles; Gonzàlez, Jordi Abstract: The increasing interest in Cognitive Vision Systems (CVS) motivates the apparition of ad-hoc stages designed for the integration of multiple kinds of knowledge. This paper proposes a novel ontology to restrict and integrate high-level semantics for Human Sequence Evaluation (HSE), which targets multilingual capabilities and multipurpose end-user interfaces. The main contributions of this paper are the conception of a neutral semantic layer, which allows to link vision and linguistic domains; and the use of situations instead of verbs as basic elements for an ontological categorization of occurrences. In our approach, the domain has been restricted to outdoor surveilled scenarios, involving interactions among pedestrians, static objects, and vehicular traffic. Fri, 13 Mar 2009 09:21:19 GMT http://hdl.handle.net/2117/2684 2009-03-13T09:21:19Z Fernández, Carles; Gonzàlez, Jordi no The increasing interest in Cognitive Vision Systems (CVS) motivates the apparition of ad-hoc stages designed for the integration of multiple kinds of knowledge. This paper proposes a novel ontology to restrict and integrate high-level semantics for Human Sequence Evaluation (HSE), which targets multilingual capabilities and multipurpose end-user interfaces. The main contributions of this paper are the conception of a neutral semantic layer, which allows to link vision and linguistic domains; and the use of situations instead of verbs as basic elements for an ontological categorization of occurrences. In our approach, the domain has been restricted to outdoor surveilled scenarios, involving interactions among pedestrians, static objects, and vehicular traffic. http://hdl.handle.net/2117/2683 Title: Unidimensional multiscale local features for object detection under rotation and mild occlusions Authors: Villamizar Vergel, Michael Alejandro; Sanfeliu Cortés, Alberto; Andrade-Cetto, Juan Abstract: In this article, scale and orientation invariant object detection is performed by matching intensity level histograms. Unlike other global measurement methods, the present one uses a local feature description that allows small changes in the histogram signature, giving robustness to partial occlusions. Local features over the object histogram are extracted during a Boosting learning phase, selecting the most discriminant features within a training histogram image set. The Integral Histogram has been used to compute local histograms in constant time. Fri, 13 Mar 2009 09:21:10 GMT http://hdl.handle.net/2117/2683 2009-03-13T09:21:10Z Villamizar Vergel, Michael Alejandro; Sanfeliu Cortés, Alberto; Andrade-Cetto, Juan no In this article, scale and orientation invariant object detection is performed by matching intensity level histograms. Unlike other global measurement methods, the present one uses a local feature description that allows small changes in the histogram signature, giving robustness to partial occlusions. Local features over the object histogram are extracted during a Boosting learning phase, selecting the most discriminant features within a training histogram image set. The Integral Histogram has been used to compute local histograms in constant time. http://hdl.handle.net/2117/2682 Title: Robust color contour object detection invariant to shadows Authors: Scandaliaris, Jorge; Villamizar Vergel, Michael Alejandro; Andrade-Cetto, Juan; Sanfeliu Cortés, Alberto Abstract: In this work a new robust color and contour based object detection method in images with varying shadows is presented. The method relies on a physics-based contour detector that emphasizes material changes and a contour-based boosted classifier. The method has been tested in a sequence of outdoor color images presenting varying shadows using two classifiers, one that learns contour object features from a simple gradient detector, and another that learns from the photometric invariant contour detector. It is shown that the detection performance of the classifier trained with the photometric invariant detector is significantly higher than that of the classifier trained with gradient detector. Fri, 13 Mar 2009 09:21:01 GMT http://hdl.handle.net/2117/2682 2009-03-13T09:21:01Z Scandaliaris, Jorge; Villamizar Vergel, Michael Alejandro; Andrade-Cetto, Juan; Sanfeliu Cortés, Alberto no color invariance, shadow removal, object detection, boosting In this work a new robust color and contour based object detection method in images with varying shadows is presented. The method relies on a physics-based contour detector that emphasizes material changes and a contour-based boosted classifier. The method has been tested in a sequence of outdoor color images presenting varying shadows using two classifiers, one that learns contour object features from a simple gradient detector, and another that learns from the photometric invariant contour detector. It is shown that the detection performance of the classifier trained with the photometric invariant detector is significantly higher than that of the classifier trained with gradient detector.