Search

Global visual localization estimates the absolute pose of a camera using a single image, in a previously mapped area. Obtaining the pose from a single image enables many robotics and augmented/virtual reality applications. Inspired by latest advances in deep learning, many existing approaches directly learn and regress 6 DoF pose from an input image. However, these methods do not fully utilize the underlying scene geometry for pose regression. The challenge in monocular relocalization is the minimal availability of supervised training data, which is just the corresponding 6 DoF poses of the images. In this paper, we propose to utilize these minimal available labels (.i.e, poses) to learn the underlying 3D geometry of the scene and use the geometry to estimate the 6 DoF camera pose. We present a learning method that uses these pose labels and rigid alignment to learn two 3D geometric representations (\textit{X, Y, Z coordinates}) of the scene, one in camera coordinate frame and the other in global coordinate frame. Given a single image, it estimates these two 3D scene representations, which are then aligned to estimate a pose that matches the pose label. This formulation allows for the active inclusion of additional learning constraints to minimize 3D alignment errors between the two 3D scene representations, and 2D re-projection errors between the 3D global scene representation and 2D image pixels, resulting in improved localization accuracy. During inference, our model estimates the 3D scene geometry in camera and global frames and aligns them rigidly to obtain pose in real-time. We evaluate our work on three common visual localization datasets, conduct ablation studies, and show that our method exceeds state-of-the-art regression methods' pose accuracy on all datasets., Comment: IEEE ROBOTICS AND AUTOMATION LETTERS. ACCEPTED NOVEMBER, 2023

Existing data on building destruction in conflict zones rely on eyewitness reports or manual detection, which makes it generally scarce, incomplete, and potentially biased. This lack of reliable data imposes severe limitations for media reporting, humanitarian relief efforts, human-rights monitoring, reconstruction initiatives, and academic studies of violent conflict. This article introduces an automated method of measuring destruction in high-resolution satellite images using deep-learning techniques combined with label augmentation and spatial and temporal smoothing, which exploit the underlying spatial and temporal structure of destruction. As a proof of concept, we apply this method to the Syrian civil war and reconstruct the evolution of damage in major cities across the country. Our approach allows generating destruction data with unprecedented scope, resolution, and frequency-and makes use of the ever-higher frequency at which satellite imagery becomes available.

We introduce the problem of weakly supervised Multi-Object Tracking and Segmentation, i.e. joint weakly supervised instance segmentation and multi-object tracking, in which we do not provide any kind of mask annotation. To address it, we design a novel synergistic training strategy by taking advantage of multi-task learning, i.e. classification and tracking tasks guide the training of the unsupervised instance segmentation. For that purpose, we extract weak foreground localization information, provided by Grad-CAM heatmaps, to generate a partial ground truth to learn from. Additionally, RGB image level information is employed to refine the mask prediction at the edges of the objects. We evaluate our method on KITTI MOTS, the most representative benchmark for this task, reducing the performance gap on the MOTSP metric between the fully supervised and weakly supervised approach to just 12% and 12.7% for cars and pedestrians, respectively., Comment: Accepted at Autonomous Vehicle Vision WACV 2021 Workshop

[EN] Virtual reconstruction of the fortified bridge of the Media Luna (Cartagena de Indias) using digital toolsOne of the most important architectural pieces of the walled complex of Cartagena de Indias in Colombia was the so-called bridge of the Media Luna, whose first phase of construction dates back to 1628. Since its construction, the Media Luna underwent several reforms and was submitted to some projects aimed at strengthening its defensive capacity conceived by military engineers at the service of the Spanish Crown, such as Juan Jiménez Donoso and Antonio de Arévalo. Despite this, a complete plan of the bridge, demolished at the end of the nineteenth century, has not been preserved, and it was never completely photographed. This paper, as a product of a research project related to the genesis of construction techniques in Cartagena de Indias, presents a digital reconstruction of the Media Luna after gathering and analyzing cartographic information still preserved in archives, photographs taken shortly before the demolition of the bridge, as well as traveler stories, descriptions and drawings done by nineteenth century chroniclers. The integral visualization of the Media Luna and its surroundings offers new value to the bridge and contributes to the integral understanding of the fortified complex of Cartagena de Indias.

[EN] Virtual reconstruction of the fortified bridge of the Media Luna (Cartagena de Indias) using digital toolsOne of the most important architectural pieces of the walled complex of Cartagena de Indias in Colombia was the so-called bridge of the Media Luna, whose first phase of construction dates back to 1628. Since its construction, the Media Luna underwent several reforms and was submitted to some projects aimed at strengthening its defensive capacity conceived by military engineers at the service of the Spanish Crown, such as Juan Jiménez Donoso and Antonio de Arévalo. Despite this, a complete plan of the bridge, demolished at the end of the nineteenth century, has not been preserved, and it was never completely photographed. This paper, as a product of a research project related to the genesis of construction techniques in Cartagena de Indias, presents a digital reconstruction of the Media Luna after gathering and analyzing cartographic information still preserved in archives, photographs taken shortly before the demolition of the bridge, as well as traveler stories, descriptions and drawings done by nineteenth century chroniclers. The integral visualization of the Media Luna and its surroundings offers new value to the bridge and contributes to the integral understanding of the fortified complex of Cartagena de Indias.

[EN] Virtual reconstruction of the fortified bridge of the Media Luna (Cartagena de Indias) using digital toolsOne of the most important architectural pieces of the walled complex of Cartagena de Indias in Colombia was the so-called bridge of the Media Luna, whose first phase of construction dates back to 1628. Since its construction, the Media Luna underwent several reforms and was submitted to some projects aimed at strengthening its defensive capacity conceived by military engineers at the service of the Spanish Crown, such as Juan Jiménez Donoso and Antonio de Arévalo. Despite this, a complete plan of the bridge, demolished at the end of the nineteenth century, has not been preserved, and it was never completely photographed. This paper, as a product of a research project related to the genesis of construction techniques in Cartagena de Indias, presents a digital reconstruction of the Media Luna after gathering and analyzing cartographic information still preserved in archives, photographs taken shortly before the demolition of the bridge, as well as traveler stories, descriptions and drawings done by nineteenth century chroniclers. The integral visualization of the Media Luna and its surroundings offers new value to the bridge and contributes to the integral understanding of the fortified complex of Cartagena de Indias.

[EN] Virtual reconstruction of the fortified bridge of the Media Luna (Cartagena de Indias) using digital toolsOne of the most important architectural pieces of the walled complex of Cartagena de Indias in Colombia was the so-called bridge of the Media Luna, whose first phase of construction dates back to 1628. Since its construction, the Media Luna underwent several reforms and was submitted to some projects aimed at strengthening its defensive capacity conceived by military engineers at the service of the Spanish Crown, such as Juan Jiménez Donoso and Antonio de Arévalo. Despite this, a complete plan of the bridge, demolished at the end of the nineteenth century, has not been preserved, and it was never completely photographed. This paper, as a product of a research project related to the genesis of construction techniques in Cartagena de Indias, presents a digital reconstruction of the Media Luna after gathering and analyzing cartographic information still preserved in archives, photographs taken shortly before the demolition of the bridge, as well as traveler stories, descriptions and drawings done by nineteenth century chroniclers. The integral visualization of the Media Luna and its surroundings offers new value to the bridge and contributes to the integral understanding of the fortified complex of Cartagena de Indias.

[EN] Virtual reconstruction of the fortified bridge of the Media Luna (Cartagena de Indias) using digital toolsOne of the most important architectural pieces of the walled complex of Cartagena de Indias in Colombia was the so-called bridge of the Media Luna, whose first phase of construction dates back to 1628. Since its construction, the Media Luna underwent several reforms and was submitted to some projects aimed at strengthening its defensive capacity conceived by military engineers at the service of the Spanish Crown, such as Juan Jiménez Donoso and Antonio de Arévalo. Despite this, a complete plan of the bridge, demolished at the end of the nineteenth century, has not been preserved, and it was never completely photographed. This paper, as a product of a research project related to the genesis of construction techniques in Cartagena de Indias, presents a digital reconstruction of the Media Luna after gathering and analyzing cartographic information still preserved in archives, photographs taken shortly before the demolition of the bridge, as well as traveler stories, descriptions and drawings done by nineteenth century chroniclers. The integral visualization of the Media Luna and its surroundings offers new value to the bridge and contributes to the integral understanding of the fortified complex of Cartagena de Indias.

Existing data on building destruction in conflict zones rely on eyewitness reports or manual detection, which makes it generally scarce, incomplete and potentially biased. This lack of reliable data imposes severe limitations for media reporting, humanitarian relief efforts, human rights monitoring, reconstruction initiatives, and academic studies of violent conflict. This article introduces an automated method of measuring destruction in high-resolution satellite images using deep learning techniques combined with data augmentation to expand training samples. We apply this method to the Syrian civil war and reconstruct the evolution of damage in major cities across the country. The approach allows generating destruction data with unprecedented scope, resolution, and frequency - only limited by the available satellite imagery - which can alleviate data limitations decisively.

Presentació de la ponència a càrrec de Joan Fontàs Serrat de la Universitat de Girona

Seguramente coincidiremos en lo interesante que podría ser aprender construcción visitando y analizando obras reconocidas de arquitectura. No obstante, resulta complejo localizar edificios cercanos visitables dentro del periodo académico y cuyo interés dé respuesta a los contenidos de la asignatura. Ahí es cuando nos planteamos el posible aprendizaje a través del uso de fotografías del edifico aportadas por los mismos estudiantes a través de un foro virtual. En el contexto del desarrollo de la tesis "El aprendizaje de la construcción a través de las imágenes" la hipótesis es que, con este método participativo, el conocimiento se va consolidando progresivamente a través de la observación y análisis de las fotografías. Las encuestas realizadas entre los estudiantes pretenden determinar su percepción sobre el aprendizaje obtenido. El análisis de los resultados confirma que ha aumentado la motivación de los estudiantes para saber cómo están construidos los edificios y el aprendizaje es bastante significativo., Surely, one could agree on how interesting it could be to learn construction by visiting and analyzing recognized works of architecture. However, it is complex to locate nearby buildings that can be visited within the academic period and whose interest responds to the contents of the subject. That is when we consider possible learning with photographs of the building, which are provided by the students themselves through a virtual forum. In the context of the development of the thesis "Learning from Construction through Images", the hypothesis is that, with this participatory method, knowledge is gradually consolidated through the observation and analysis of photographs. Surveys conducted among students aim to determine their perception of the learning obtained. The analysis of the results confirms that have increased the students’ motivation to know how the buildings are constructed and their learning is quite significant., Peer Reviewed

In this work we contribute a novel pipeline to automatically generate training data, and to improve over state-of-the-art multi-object tracking and segmentation (MOTS) methods. Our proposed track mining algorithm turns raw street-level videos into high-fidelity MOTS training data, is scalable and overcomes the need of expensive and time-consuming manual annotation approaches. We leverage state-of-the-art instance segmentation results in combination with optical flow predictions, also trained on automatically harvested training data. Our second major contribution is MOTSNet - a deep learning, tracking-by-detection architecture for MOTS - deploying a novel mask-pooling layer for improved object association over time. Training MOTSNet with our automatically extracted data leads to significantly improved sMOTSA scores on the novel KITTI MOTS dataset (+1.9%/+7.5% on cars/pedestrians), and MOTSNet improves by +4.1% over previously best methods on the MOTSChallenge dataset. Our most impressive finding is that we can improve over previous best-performing works, even in complete absence of manually annotated MOTS training data.

The Hierarchical Conditional Random Field (HCRF) model have been successfully applied to a number of image labeling problems, including image segmentation. However, existing HCRF models of image segmentation do not allow multiple classes to be assigned to a single region, which limits their ability to incorporate contextual information across multiple scales. At higher scales in the image, this representation yields an oversimplified model since multiple classes can be reasonably expected to appear within large regions. This simplified model particularly limits the impact of information at higher scales. Since class-label information at these scales is usually more reliable than at lower, noisier scales, neglecting this information is undesirable. To address these issues, we propose a new consistency potential for image labeling problems, which we call the harmony potential. It can encode any possible combination of labels, penalizing only unlikely combinations of classes. We also propose an effective sampling strategy over this expanded label set that renders tractable the underlying optimization problem. Our approach obtains state-of-the-art results on two challenging, standard benchmark datasets for semantic image segmentation: PASCAL VOC 2010, and MSRC-21

When neural networks process images which do not resemble the distribution seen during training, so called out-of-distribution images, they often make wrong predictions, and do so too confidently. The capability to detect out-of-distribution images is therefore crucial for many real-world applications. We divide out-of-distribution detection between novelty detection ---images of classes which are not in the training set but are related to those---, and anomaly detection ---images with classes which are unrelated to the training set. By related we mean they contain the same type of objects, like digits in MNIST and SVHN. Most existing work has focused on anomaly detection, and has addressed this problem considering networks trained with the cross-entropy loss. Differently from them, we propose to use metric learning which does not have the drawback of the softmax layer (inherent to cross-entropy methods), which forces the network to divide its prediction power over the learned classes. We perform extensive experiments and evaluate both novelty and anomaly detection, even in a relevant application such as traffic sign recognition, obtaining comparable or better results than previous works., Comment: Accepted at BMVC 2018, 10 pages main article and 4 pages supplementary material

We are concerned with whether a vertically integrated broadband and content provider can unreasonably advantage itself over competing content providers, either by selling quality-of-service (QoS) to content providers at unreasonably high prices, or by refusing to provide access to QoS to competing content. We address this question by modeling the competition between one such vertically integrated provider and one over-the-top (OTT) content provider. We analytically determine when the broadband provider will sell QoS and when the OTT content provider or users will purchase QoS. We characterize the optimal QoS and video service prices. The ISP's market share increases with the difference in the value of the two video services and decreases with the difference in the corresponding costs. Numerical results illustrate the effect of QoS price on content price, and the variation of market share and profit with QoS price. The ISP may sell QoS to users at a lower price than when QoS is sold to the OTT provider.

We are concerned with whether a vertically integrated broadband and content provider can unreasonably advantage itself over competing content providers, either by selling quality-of-service (QoS) to content providers at unreasonably high prices, or by refusing to provide access to QoS to competing content. We address this question by modeling the competition between one such vertically integrated provider and one over-the-top (OTT) content provider. We analytically determine when the broadband provider will sell QoS and when the OTT content provider or users will purchase QoS. We characterize the optimal QoS and video service prices. The ISP's market share increases with the difference in the value of the two video services and decreases with the difference in the corresponding costs. Numerical results illustrate the effect of QoS price on content price, and the variation of market share and profit with QoS price. The ISP may sell QoS to users at a lower price than when QoS is sold to the OTT provider.

Network Function Virtualization (NFV) has drawn significant attention from both industry and academia as an important shift in telecommunication service provisioning. By decoupling Network Functions (NFs) from the physical devices on which they run, NFV has the potential to lead to significant reductions in Operating Expenses (OPEX) and Capital Expenses (CAPEX) and facilitate the deployment of new services with increased agility and faster time-to-value. The NFV paradigm is still in its infancy and there is a large spectrum of opportunities for the research community to develop new architectures, systems and applications, and to evaluate alternatives and trade-offs in developing technologies for its successful deployment. In this paper, after discussing NFV and its relationship with complementary fields of Software Defined Networking (SDN) and cloud computing, we survey the state-of-the-art in NFV, and identify promising research directions in this area. We also overview key NFV projects, standardization efforts, early implementations, use cases and commercial products., Comment: 28 Pages in IEEE Communications Surveys and Tutorials. September 2015

As the virtualization of networks continues to attract attention from both industry and academia, the Virtual Network Embedding (VNE) problem remains a focus of researchers. This paper proposes a one-shot, unsplittable flow VNE solution based on column generation. We start by formulating the problem as a path-based mathematical program called the primal, for which we derive the corresponding dual problem. We then propose an initial solution which is used, first, by the dual problem and then by the primal problem to obtain a final solution. Unlike most approaches, our focus is not only on embedding accuracy but also on the scalability of the solution. In particular, the one-shot nature of our formulation ensures embedding accuracy, while the use of column generation is aimed at enhancing the computation time to make the approach more scalable. In order to assess the performance of the proposed solution, we compare it against four state of the art approaches as well as the optimal link-based formulation of the one-shot embedding problem. Experiments on a large mix of Virtual Network (VN) requests show that our solution is near optimal (achieving about 95% of the acceptance ratio of the optimal solution), with a clear improvement over existing approaches in terms of VN acceptance ratio and average Substrate Network (SN) resource utilization, and a considerable improvement (92% for a SN of 50 nodes) in time complexity compared to the optimal solution., Comment: 14 Pages in IEEE Transactions on Network and Service Management, September 2015

Virtualization enables the building of multiple virtual networks over a shared substrate. One of the challenges to virtualisation is efficient resource allocation. This problem has been found to be NP hard. Therefore, most approaches to it have not only proposed static solutions, but have also made many assumptions to simplify it. In this paper, we propose a distributed, autonomic and artificial intelligence based solution to resource allocation. Our aim is to obtain self-configuring, selfoptimizing, self-healing and context aware virtual networks, Comment: Short Paper, 4 Pages, Summer School, PhD Work In Progress Workshop. Scalable and Adaptive Internet Solutions (SAIL). June 2012

A new era in telecommunications is emerging. Virtualized networking functions and resources will offer network operators a way to shift the balance of expenditure from capital to operational, opening up networks to new and innovative services. This article introduces the concept of edge as a service (EaaS), a means of harnessing the flexibility of virtualized network functions and resources to enable network operators to break the tightly coupled relationship they have with their infrastructure and enable more effective ways of generating revenue. To achieve this vision, we envisage a virtualized service access interface that can be used to programmatically alter access network functions and resources available to service providers in an elastic fashion. EaaS has many technically and economically difficult challenges that must be addressed before it can become a reality; the main challenges are summarized in this article.

Vision-based road detection is an essential functionality for supporting advanced driver assistance systems (ADAS) such as road following and vehicle and pedestrian detection. The major challenges of road detection are dealing with shadows and lighting variations and the presence of other objects in the scene. Current road detection algorithms characterize road areas at pixel level and group pixels accordingly. However, these algorithms fail in presence of strong shadows and lighting variations. Therefore, we propose a road detection algorithm based on video alignment. The key idea of the algorithm is to exploit the similarities occurred when a vehicle follows the same trajectory more than once. In this way, road areas are learned in a first ride and then, this road knowledge is used to infer areas depicting drivable road surfaces in subsequent rides. Two different experiments are conducted to validate the proposal on different video sequences taken at different scenarios and different daytime. The former aims to perform on-line road detection. The latter aims to perform off-line road detection and is applied to automatically generate the ground-truth necessary to validate road detection algorithms. Qualitative and quantitative evaluations prove that the proposed algorithm is a valid road detection approach.

The Autonomic Internet project [1] approach relies on abstractions and distributed systems of a five plane solution for the provision of Future Internet Services (OSKMV): Orchestration, Service Enablers, Knowledge, Management and Virtualisation Planes. This paper presents a practical viewpoint of the manageability of virtual networks, exercising the components and systems that integrate this approach and that are being validated. This paper positions the distributed systems and networking services that integrate this solution, focusing on the provision of Future Internet services for self-configuration and selfperformance management scenes.

El presente anteproyecto de “Desdoblamiento del tramo final de la línea Castellbisbal-Mollet del pK 23,2 al pK 27,3 y nueva estación de Mollet-Sant Fost” nace con el objetivo de llevar a cabo las actuaciones necesarias para dotar a una infraestructura básica como es la línea de ferrocarril Castellbisbal-Mollet de doble vía en toda su traza. Actualmente, la ausencia de doble vía en el tramo final causa graves interferencias con el servicio del Corredor del Mediterráneo y la línea de cercanías C2 Vilanova – Barcelona - Maçanet, al existir cizallamientos en las vías principales.

The paper is about PROTON, a reactive policy management system developed to set up policies to decide when vertical handover should occur, based on a 3-layer subsystem. This work is significant because it showed that PROTON could not be run on the mobile node. The paper proposed running PROTON in the network that would then specify a finite-state machine (i.e. an autonomic system) to execute on the mobile. The work is significant because it showed that autonomic techniques can be used in 4G systems to support client-based handover. It is also a tremendously useful paper for implementers of real systems.

The paper is about PROTON, a reactive policy management system developed to set up policies to decide when vertical handover should occur, based on a 3-layer subsystem. This work is significant because it showed that PROTON could not be run on the mobile node. The paper proposed running PROTON in the network that would then specify a finite-state machine (i.e. an autonomic system) to execute on the mobile. The work is significant because it showed that autonomic techniques can be used in 4G systems to support client-based handover. It is also a tremendously useful paper for implementers of real systems.

The paper is about PROTON, a reactive policy management system developed to set up policies to decide when vertical handover should occur, based on a 3-layer subsystem. This work is significant because it showed that PROTON could not be run on the mobile node. The paper proposed running PROTON in the network that would then specify a finite-state machine (i.e. an autonomic system) to execute on the mobile. The work is significant because it showed that autonomic techniques can be used in 4G systems to support client-based handover. It is also a tremendously useful paper for implementers of real systems.

Intérpretes: Antonio Carrizo. Horacio Carballal. Alberto Cortés. Julio Díaz Usandivaras. Alfredo Sánchez Bella. Los Waldos. Joan Manuel Serrat. Manuel González Haba

The Hierarchical Conditional Random Field (HCRF) model have been successfully applied to a number of image labeling problems, including image segmentation. However, existing HCRF models of image segmentation do not allow multiple classes to be assigned to a single region, which limits their ability to incorporate contextual information across multiple scales. At higher scales in the image, this representation yields an oversimplified model since multiple classes can be reasonably expected to appear within large regions. This simplified model particularly limits the impact of information at higher scales. Since class-label information at these scales is usually more reliable than at lower, noisier scales, neglecting this information is undesirable. To address these issues, we propose a new consistency potential for image labeling problems, which we call the harmony potential. It can encode any possible combination of labels, penalizing only unlikely combinations of classes. We also propose an effective sampling strategy over this expanded label set that renders tractable the underlying optimization problem. Our approach obtains state-of-the-art results on two challenging, standard benchmark datasets for semantic image segmentation: PASCAL VOC 2010, and MSRC-21