Capítols de llibre
http://hdl.handle.net/2117/3757
2016-05-26T01:31:00ZDense segmentation-aware descriptors
http://hdl.handle.net/2117/85171
Dense segmentation-aware descriptors
Trulls Fortuny, Eduard; Kokkinos, Iasonas; Sanfeliu Cortés, Alberto; Moreno-Noguer, Francesc
Dense descriptors are becoming increasingly popular in a host of tasks, such as dense image correspondence, bag-of-words image classification, and label transfer. However, the extraction of descriptors on generic image points, rather than selecting geometric features, requires rethinking how to achieve invariance to nuisance parameters. In this work we pursue invariance to occlusions and background changes by introducing segmentation information within dense feature construction. The core idea is to use the segmentation cues to downplay the features coming from image areas that are unlikely to belong to the same region as the feature point. We show how to integrate this idea with dense SIFT, as well as with the dense scale- and rotation-invariant descriptor (SID). We thereby deliver dense descriptors that are invariant to background changes, rotation, and/or scaling. We explore the merit of our technique in conjunction with large displacement motion estimation and wide-baseline stereo, and demonstrate that exploiting segmentation information yields clear improvements.
2016-04-05T08:27:39ZTrulls Fortuny, EduardKokkinos, IasonasSanfeliu Cortés, AlbertoMoreno-Noguer, FrancescDense descriptors are becoming increasingly popular in a host of tasks, such as dense image correspondence, bag-of-words image classification, and label transfer. However, the extraction of descriptors on generic image points, rather than selecting geometric features, requires rethinking how to achieve invariance to nuisance parameters. In this work we pursue invariance to occlusions and background changes by introducing segmentation information within dense feature construction. The core idea is to use the segmentation cues to downplay the features coming from image areas that are unlikely to belong to the same region as the feature point. We show how to integrate this idea with dense SIFT, as well as with the dense scale- and rotation-invariant descriptor (SID). We thereby deliver dense descriptors that are invariant to background changes, rotation, and/or scaling. We explore the merit of our technique in conjunction with large displacement motion estimation and wide-baseline stereo, and demonstrate that exploiting segmentation information yields clear improvements.Robot interactive learning through human assistance
http://hdl.handle.net/2117/23219
Robot interactive learning through human assistance
Ferrer Mínguez, Gonzalo; Garrell Zulueta, Anais; Villamizar Vergel, Michael Alejandro; Huerta Casado, Iván; Sanfeliu Cortés, Alberto
2014-06-13T13:27:59ZFerrer Mínguez, GonzaloGarrell Zulueta, AnaisVillamizar Vergel, Michael AlejandroHuerta Casado, IvánSanfeliu Cortés, AlbertoDebats via Twitter
http://hdl.handle.net/2117/23158
Debats via Twitter
Domingo Peña, Joan; Segura Casanovas, Joan; Durán Moyano, José Luis; Grau Saldes, Antoni; Bolea Monte, Yolanda
Es presenta una experiència de debat entre estudiants, tutelat pel professorat. Els
debats a l’aula són una metodologia activa que, malgrat les seves virtuts, habitualment genera poca participació i que està força concentrada en uns quants estudiants. En aquest fet influeixen diversos factors i el resultat global no és sempre
satisfactori. L'experiència que s'ha realitzat consisteix en debats mitjançant una xarxa social al llarg d'uns quants dies cada curs. Amb això s'obté una elevada participació, propera al 100 %, alhora que d'aquest debat deriven situacions que es poden tractar, ja en temps presencial, de forma més completa.
S'ha realitzat aquesta experiència al llarg de dos cursos amb un total d'uns 450 estudiants i el resultat ha estat altament satisfactori. El text que segueix exposa les bases d'aquest sistema de treball així, com els elements per fer-ne l’anàlisi.
2014-06-04T12:26:07ZDomingo Peña, JoanSegura Casanovas, JoanDurán Moyano, José LuisGrau Saldes, AntoniBolea Monte, YolandaEs presenta una experiència de debat entre estudiants, tutelat pel professorat. Els
debats a l’aula són una metodologia activa que, malgrat les seves virtuts, habitualment genera poca participació i que està força concentrada en uns quants estudiants. En aquest fet influeixen diversos factors i el resultat global no és sempre
satisfactori. L'experiència que s'ha realitzat consisteix en debats mitjançant una xarxa social al llarg d'uns quants dies cada curs. Amb això s'obté una elevada participació, propera al 100 %, alhora que d'aquest debat deriven situacions que es poden tractar, ja en temps presencial, de forma més completa.
S'ha realitzat aquesta experiència al llarg de dos cursos amb un total d'uns 450 estudiants i el resultat ha estat altament satisfactori. El text que segueix exposa les bases d'aquest sistema de treball així, com els elements per fer-ne l’anàlisi.PLC Control and Matlab/Simulink Simulations – A Translation Approach
http://hdl.handle.net/2117/11613
PLC Control and Matlab/Simulink Simulations – A Translation Approach
Martins, Joao; Lima, Celson; Grau Saldes, Antoni; Martínez García, Herminio
2011-03-01T18:03:48ZMartins, JoaoLima, CelsonGrau Saldes, AntoniMartínez García, HerminioEnginyeria Tècnica Industrial, Especialitats en Electricitat, Electrònica, Mecànica i Química, a l'Escola Universitària d'Enginyeria Tècnica Industrial de Barcelona de la Universitat Politècnica de Catalunya
http://hdl.handle.net/2117/9689
Enginyeria Tècnica Industrial, Especialitats en Electricitat, Electrònica, Mecànica i Química, a l'Escola Universitària d'Enginyeria Tècnica Industrial de Barcelona de la Universitat Politècnica de Catalunya
Martínez García, Herminio; Jorba Peiró, Jordi; Gámiz Caro, Juan
2010-10-14T10:20:38ZMartínez García, HerminioJorba Peiró, JordiGámiz Caro, JuanLocal boosted features for pedestrian detection
http://hdl.handle.net/2117/9181
Local boosted features for pedestrian detection
Villamizar Vergel, Michael Alejandro; Sanfeliu Cortés, Alberto; Andrade-Cetto, Juan
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
2010-09-29T17:15:00ZVillamizar Vergel, Michael AlejandroSanfeliu Cortés, AlbertoAndrade-Cetto, JuanThe 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 ratesA recursive embedding approach to median graph computation
http://hdl.handle.net/2117/7811
A recursive embedding approach to median graph computation
Ferrer Sumsi, Miquel; Valveny, Ernest; Bunke, Horst
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.
2010-06-23T10:24:44ZFerrer Sumsi, MiquelValveny, ErnestBunke, HorstThe 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.Graph-based k-means clustering: A comparison of the set versus the generalized median graph
http://hdl.handle.net/2117/7485
Graph-based k-means clustering: A comparison of the set versus the generalized median graph
Ferrer Sumsi, Miquel; Valveny, Ernest; Serratosa Casanelles, Francesc; Bardaji Goikoetxea, Itziar; Bunke, Horst
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.
2010-06-02T12:49:21ZFerrer Sumsi, MiquelValveny, ErnestSerratosa Casanelles, FrancescBardaji Goikoetxea, ItziarBunke, HorstIn 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.Median graph computation by means of a genetic approach based on minimum common supergraph and maximum common subraph
http://hdl.handle.net/2117/7107
Median graph computation by means of a genetic approach based on minimum common supergraph and maximum common subraph
Ferrer Sumsi, Miquel; Valveny, Ernest; Serratosa Casanelles, Francesc
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.
2010-05-03T13:01:33ZFerrer Sumsi, MiquelValveny, ErnestSerratosa Casanelles, FrancescGiven 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.Unidimensional multiscale local features for object detection under rotation and mild occlusions
http://hdl.handle.net/2117/2683
Unidimensional multiscale local features for object detection under rotation and mild occlusions
Villamizar Vergel, Michael Alejandro; Sanfeliu Cortés, Alberto; Andrade-Cetto, Juan
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.
2009-03-13T09:21:10ZVillamizar Vergel, Michael AlejandroSanfeliu Cortés, AlbertoAndrade-Cetto, JuanIn 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.