RecPad: repositório

201620152014201320122011201020092007200820022000

RecPad 2016

22th edition of the Portuguese Conference On Pattern Recognition
Esta edição do Recpad foi organizada por um comité da Universidade de Aveiro e decorreu na Universidade de Aveiro.
Comité organizativo
  • Armando J. Pinho (UA)
  • Diogo Pratas (UA)
  • Raquel Sebastião (UA)
  • Samuel Silva (UA)
  • Sónia Gouveia (UA)
  • Susana Brás (UA)
Portuguese Association for Pattern Recognition  

Orador convidado

OLYMPUS DIGITAL CAMERA

Bio

Joan Serra-Sagristà (IEEE Senior Member 2011) received the Ph.D. degree in computer science from Universitat Autònoma de Barcelona (UAB), Spain, in 1999. He is currently an Associate Professor at Department of Information and Communications Engineering, UAB. He holds the Accreditation as Full Professor from both Spanish ANECA and Catalan AQU Catalunya. From September 1997 to December 1998, he was at University of Bonn, Germany, funded by DAAD. His current research interests focus on data compression, with special attention to image coding for remote sensing and telemedicine applications. He serves as Associate Editor of IEEE Trans. on Image Processing and as Program Committee co-chair for IEEE Data Compression Conference. He has co-authored over one hundred publications. He was the recipient of the Spanish Intensification Young Investigator Award in 2006.

Remote sensing data compression

This talk describes recent developments in several areas of remote sensing data compression. The first part of the talk will introduce the current status of Earth Observation missions and the need for efficient data transmission, where data compression plays a significant role. The second part of the talk will be dedicated to onboard compression of remote sensing data, in particular to recent and ongoing work developed by the main space agencies. The third part of the talk will introduce some of our own recent developments in this field.

 

Local
UA - Universidade de Aveiro
Data
10 October 2016
Website
http://recpad2016.web.ua.pt
Logo
Could not load cover

RecPad 2015

21th edition of the Portuguese Conference On Pattern Recognition
Esta edição do Recpad foi organizada por um comité da Universidade do Algarve e decorreu no Instituto Superior de Engenharia da Universidade do Algarve.
Comité organizativo
Universidade do Algarve Portuguese Association for Pattern Recognition Cintal  
Patrocinadores
Foi patrocinado pela SPIC e pelo Hotel EVA. SPIC Hotel EVA

Orador convidado

norbert_kruger

 

Norbert Krüger

Maersk MC-Kinney Moller Institute for Production Technology,

Technical Faculty at the University of Southern Denmark.

Denmark

 

Bio

Norbert Krüger has been employed at the University of Southern Denmark since 2006 (first as an Associate Professor and then as a full Professor (MSO) since 2008). He is one of the two leaders of the Cognitive and Applied Robotics Group (CARO, caro.sdu.dk) in which currently 12 PhD students, two Assistant and two Associate Professor as well as 8 master students are working. Norbert Krüger’s research focuses on Cognitive Vision, in particular vision based manipulation and learning. He has published 45 papers in journals and more than 80 papers at conferences covering the topics computer vision, robotics, neuroscience as well as cognitive systems. His H-index is 24. His group has developed the C++-software CoViS (Cognitive Vision Software) which is now used by a number of groups in national as well as European projects. He is currently involved in 2 European projects as well as 4 Danish projects.

 

Deep Hierarchies in Human and Computer Vision

Computer vision – although being still a rather young scientific discipline – in the last decades was able to provide some impressive examples of artificial vision systems that outperform humans. However, the human visual system is still superior to any artificial vision system in visual tasks requiring generalization and reasoning (often also called ‘cognitive vision’) such as extraction of visual based affordances or visual tasks in the context of tool use and dexterous manipulation of unknown objects.

Two decades ago, there has been a strong connection between the two communities dealing with human vision research and computer vision. This link however has been somehow lost recently and computer vision has been more and more developed into a sub-field of machine learning. In this talk, I argue that the reason for the superiority of human vision for ‘cognitive vision tasks’ is connected to the deep hierarchical architecture of the primate’s visual system.

The talk is divided into two parts: First, I will give an overview about today’s knowledge about the primate’s (and by that the human’s) visual system primarily based on neurophysiological research. This part is based on the paper (Kruger et al. 2013, IEEE PAMI) and is in particular addressing computer vision and machine learning scientists as audience.

In the second part of the talk, I will describe a three level hierarchical cognitive robot system in which actions are learned by observing humans performing these actions (Kruger et al. 2013, KI). Learning is taking place at the different levels of the hierarchy in rather different representations. On the sensory-motor level, the shape and appearance of objects as well as optimal action trajectories and force torque profiles are represented. On the mid-level, a discrete visual representation based on semantic event chains (Aksoy et al. 2011) bridges towards the planning level, the highest representational level. I will describe different the learning problems on the different representational levels and their interaction.

 

Local
ISE - Universidade do Algarve
Data
9 September 2015
Website
http://recpad2015.ualg.pt/
Logo
Could not load cover
Download

RecPad 2014

20th edition of the Portuguese Conference on Pattern Recognition.
Esta edição do Recpad foi organizada por um comitté do University of Beira Interior e decorreu no Museu Royal Veiga Factory.
Comité organizativo
University of Beira Interior   Portuguese Association for Pattern Recognition   Soft Computing and Image Analysis Group  
Patrocinadores
Foi patrocinado pela EyeSee Lda e pela INDRA PortugalInova Prime.   EyeSee Solutions    INDRA Portugal     Inova Prime  

Orador convidado

francesc_moreno

 

Associate Researcher
Institut de Robòtica i Informàtica Industrial (CSIC-UPC)
Barcelona, Spain

Bio

Francesc Moreno-Noguer received the MSc degrees in industrial engineering and electronics from the Technical University of Catalonia (UPC) and the Universitat de Barcelona in 2001 and 2002, respectively, and the PhD degree from UPC in 2005. From 2006 to 2008, he was a postdoctoral fellow at the computer vision departments of Columbia University and the Ecole Polytecnique Fédérale de Lausanne. In 2009, he joined the Institut de Robòtica i Informàtica Industrial in Barcelona as an associate researcher of the Spanish Scientific Research Council. His research interests include retrieving rigid and nonrigid shape, motion, and camera pose from single images and video sequences, with applications to both robotics and medical imaging. He received UPC’s Doctoral Dissertation Extraordinary Award for his work and an outstanding reviewer award at ECCV’12 and CVPR’14. Further information can be found at http://www.iri.upc.edu/people/fmoreno/.

Monocular 3D detection of rigid and non-rigid shapes

In this talk, I will first present an approach to the PnP problem, the estimation of the pose of a calibrated camera from n point correspondences between an image and a 3D model of a rigid object, whose computational complexity grows linearly with n. Our central idea is to express the 3D points as a weighted sum of four virtual control points. The problem then reduces to estimating the coordinates of these control points in the camera coordinate system, which can be done in O(n) using simple linearization techniques. I will then show how an algebraic outlier rejection scheme can be introduced within the computation of the pose, without the need to resort to RANSAC-based strategies.

In the second part of the talk I will show how the same linear formulations can be extended to retrieving the shape of 3D deformable objects. However, since monocular non-rigid reconstruction is severely under-constrained we will have to consider additional constraints, either based on local rigidity (to reconstruct deformable and inextensible surfaces), or based on shading coherence (to reconstruct deformable and stretchable surfaces). Finally I will discuss the major limitations of these linear formulations and propose a novel and alternative stochastic exploration strategy. I will present results both for non-rigid shape and human pose recovery.

Local
Royal Veiga Factory
Data
10 October 2014
Website
http://recpad2014.di.ubi.pt/
Logo
Could not load cover
Download

RecPad 2013

19th edition of the Portuguese Conference on Pattern Recognition.
Esta edição do Recpad foi organizada por um comitté do Instituto Superior Técnico
  • João Sanches - General chair (ISR / IST)
  • Manya Afonso - Local chair (ISR / IST)
  • David Afonso – Informatics (ISR / IST)
  • Alexandre Domingues (ISR / IST)
  • Martina Fonseca (ISR / IST)
  • Anastasiya Strembitska (ISR / IST)
Logo Instituto Superior Técnico

Orador convidado

Ana Fred

Bio

Ana L.N. Fred received the MS and PhD degrees in electrical and computer engineering, in 1989 and 1994, respectively, both from the Instituto Superior Técnico (IST), Technical University of Lisbon, Portugal.

She has been a faculty member of IST since 1986, where she is currently a professor (Professora associada com agregação) with the Department of Electrical and Computer Engineering.

She is a researcher at the Communication Theory and Pattern Recognition Group of the Institute of Telecommunications. Her research interests include information theory, pattern recognition, signal processing, and artificial intelligence.

Título:

PHYSIOLOGICAL COMPUTING – A PATTERN RECOGNITION PERSPECTIVE

Local
Instituto Superior Técnico, Lisboa
Data
11 November 2013
Website
http://soma.isr.ist.utl.pt/recpad/
Logo
Could not load cover
Download

RecPad 2012

18th edition of the Portuguese Conference on Pattern Recognition.

RecPad was organized by:

Organizing Committee

  • Nuno Cid Martins
  • Verónica Vasconcelos
  • Fernando Lopes
  • Inácio Fonseca
  • Jorge Barbosa
  • Nuno Rodrigues
  • Simão Paredes
Local Technical Support
  • Inês Duarte
  • Teresa Jorge

Orador convidado

Carlos Gonzalez-MorcilloBio

Carlos Gonzalez-Morcillo is an Associate Professor of Computer Science at the University of Castilla-La Mancha (Spain). He received the BsC and PhD degrees in Computer Science from the University of Castilla-La Mancha in 2002 and 2007, respectively. His research interests include Distributed Rendering, Augmented Reality, MultiAgent Systems, and Intelligent Surveillance. Dr Morcillo has worked in the fields of Computer Vision and Data Mining at the Software Competence Center Hagenberg (Austria). He is Blender Foundation Certified Trainer and member of the Eurographics Society. He is also co-author of the Spanish book Fundamentals of 3D Image Synthesis, a practical approach with Blender. Further information can be found at http://www.esi.uclm.es/www/cglez

Indoor Navigation Infrastructure based on Augmented Reality Techniques

Indoor navigation systems have represented a hot research topic for the research community with many different proposals to position people and keep track of their movements. In this lecture, Dr. Morcillo will take a step forward describing an object-oriented distributed architecture for highly scalable indoor navigation systems. The idea behind this architecture is to assist people with different needs while they stay in large spaces or buildings, such as public administration buildings, transport facilities, hospitals and so on, helping them reach their goals in such environments.

In this context, the use of Augmented Reality techniques boosts the concept of mobility thanks to the underlying architecture, integrating heterogeneous hardware devices and tracking methods. The resulting platform is a multi-layered scalable architecture based on autonomous agents. The most relevant work carried out by the Applied Artificial Intelligence Research Group at the University of Castilla-La Mancha will be also discussed.

Local
Coimbra
Data
10 October 2012
Website
http://www.isec.pt/recpad2012/
Logo
Could not load cover
Download

RecPad 2011

17th edition of the Portuguese Conference on Pattern Recognition
Organizing Committee Local Technical Support

Orador convidado

Daniel Gatica-PerezBio

Daniel Gatica-Perez is a senior researcher at Idiap Research Institute and the Swiss Federal Institute of Technology in Lausanne (EPFL), Switzerland, where he directs the Social Computing Group. His research integrates methods from multimedia signal processing, machine learning, ubiquitous computing, and social sciences to develop computational models for human and social behavior analysis from sensor data. His recent work has studied small groups at work in multisensor spaces, populations of smartphone users in urban environments, and on-line communities in social media. His work
has been supported by the Swiss and US governments, the European Union, and industry. Among several professional activities, he currently serves as Associate Editor of the IEEE Transactions on Multimedia, Image and Vision Computing, and the Journal of Ambient Intelligence and Smart Environments.

Reality Mining for Real: Large-Scale Human Behavior and Smartphone Data

The large-scale understanding of personal and social behavior from smartphone sensor data is an emerging trend in computing. Smartphones can constantly sense human location, motion, proximity, and communication, and represent one of the most accurate means of tracing human activities. All this data,as never before, is being generated at massive scales.

I will present an overview of recent work in my research group in this domain, which is addressing mobile sensing, data analysis, and applications. I will first describe our experience with the collection of a rich corpus of real life data using smartphones as sensors, and discuss a few of the many associated challenges – both human and technological. I will then present computational methods that we have developed to discover a variety of patterns, including daily routines of individuals, trends of phone application usage, social interaction types, and personality traits. I will finally discuss about open issues in this domain.

Local
Casa da Música, Porto
Data
10 October 2011
Website
http://paginas.fe.up.pt/~recpad2011
Logo
Could not load cover
Download

RecPad 2010

16th Portuguese Conference on Pattern Recognition
Organizing Committee
  • Paulo Salgado
  • João A. Pavão
  • Pedro Couto

Orador convidado

chamorro< Inserir foto com CCS class de “conference-guest-img” >

Bio

Jesús Chamorro Martinez – jesus@decsai.ugr.es  | Associate professor

Department of Computer Science and Artificial Intelligence – University of Granada

Jesús Chamorro was born in 1972. He received the MS and PhD degrees, both in Computer Science, from the University of Granada in 1995 and 2001 respectively. Since 1996, he is member of the Computer Science Department at the University of Granada where he is now an Associate Professor. His research interests include motion analysis and optical flow estimation, information retrieval from image databases, soft computing image processing, and biomedical image analysis. He has participated in several projects from the Spanish Research Council (CICYT) and currently he is participating in the Spanish FIS project “Imagen Médica Molecular y Multimodalidad” (IM3).

Multiple Motion Segmentation based on Spatio-temporal Volume Recognition. Application to Optical flow estimation

In this lecture, a new methodology for extracting motion patterns is applied to optical flow estimation in the presence of multiple motions. The proposed approximation deals with the problem in two stages. In the first one, the most important motions are segmented; in the second one, the optical flow is estimated on the basis of the motions detected in the previous stage. To extract relevant motions, a new approach based on a spatio-temporal filtering is presented. The approach groups together parts of a moving object that have been separated into various filter responses because of the object’s spatial structure, thereby avoiding the spatial dependency problem associated with a representation based on spatio-temporal filters. The proposed model, therefore, generates one “motion pattern” for each motion detected in the sequence. To obtain an optical flow estimation, which is able to represent multiple velocities, the gradient constraint is applied to the output of each filter so that multiple estimations of the velocity at the same location may be obtained. For each “motion pattern” detected in the previous stage, the velocities at a given point corresponding to the same motion are then combined using a probabilistic approach. In the application to optical flow estimation, the use of “motion patterns” allows multiple velocities to be represented, while the combination of estimations fromdifferent filters helps reduce the aperture problem. This technique is illustrated on real and simulated data sets, including sequences with occlusion and transparencies.

Local
Vila Real
Data
4 April 2010
Website
http://recpad2010.utad.pt/
Logo
Could not load cover

RecPad 2009

15th Portuguese Conference on Pattern Recognition

Orador convidado

 

RecPad 2007

13ª Conferência Portuguesa de Reconhecimento de Padrões
Comissão Organizadora
  • Pedro Pina (Chair, IST)
  • José Saraiva (IST)
  • Cristina Lira (IST)
  • Michele Mengucci (IST)
  • Nuno Benavente (IST)
  • Lourenço Bandeira (IST)
Comissão de Programa 
  • Aurélio Campilho (INEB/FEUP)
  • Beatriz Sousa Santos (Univ. Aveiro)
  • Helder Araújo (Univ. Coimbra)
  • João Rogério Caldas Pinto (IST)
  • Pedro Pina (IST)

Orador convidado

julio cesarJúlio Martín-Herrero, Professor da Escuela Técnica Superior de Ingenieros de Telecomunicación da Universidade de Vigo em Espanha, que tem como principais interesses de investigação o Processamento de Imagem e Visão Artificial, a Análise Espacial e as suas aplicações multidisciplinares. O título da apresentação que irá efectuar é o seguinte:

Robust Oil Slick Identification in Synthetic Aperture Radar Images with Support Vector Machines

 

Local
Lisboa
Data
10 October 2007
Website
http://cerena.ist.utl.pt/recpad2007.htm
Logo
Could not load cover
Download

RecPad 2008

14ª Conferência Portuguesa de Reconhecimento de Padrões
Comissão Organizadora
  • Jorge Batista (Chair,ISR/FCTUC)
  • Helder Araújo (ISR/FCTUC)
  • Paulo Peixoto (ISR/FCTUC)
  • Gonçalo Monteiro (ISR)
  • Pedro Martins (ISR)

Orador convidado

jordi gonzalesDr. Jordi Gonzàlez, Investigador no Institut de Robòtica i Informàtica Industrial (UPC-CSIC), Espanha.

Dr. Jordi Gonzàlez obtained his PhD degree in 2004, from Universitat Autònoma de Barcelona. At present he is a Juan de la Cierva postdoctoral researcher at the Institut de Robòtica i Informàtica Industrial (UPC-CSIC). The topic of his research is the cognitive evaluation of human behaviours in image sequences. The aim is the generation of both linguistic descriptions and virtual environments, which explain those observed behaviours. He has more than 70 publications about his research on active camera control; segmentation and tracking; human action recognition; human behaviour understanding and interpretation; natural language text generation; and automatic behavioural animation. He has also participated as a WP-leader in the European projects HERMES and VIDI-Video, and as a member in the euCognition network. He co-founded the Image Sequence Evaluation research group at the CVC in Barcelona.

Towards Human Sequence Evaluation

During the past three decades, important research efforts in computer vision have been focused on developing theories, methods and systems applied to the description of human movements in image sequences. Broadly speaking, in the past the main goal was the estimation of quantitative parameters describing where was motion. Nowadays, the focus is on the analysis of image sequences incorporating cognitive processes which allow to understand recorded movements. That is, the true challenge is the generation of qualitative descriptions about the meaning of motion, therefore understanding not only where, but also why motion is being observed. Towards this end, very few Cognitive Vision Systems (CVS) have been proposed in the literature, which typically encompass topics not only related to computer vision, but also to artificial intelligence and computational linguistics. In this talk we will restrict cognition to assure the generation of plausible semantic interpretations of human movements observed in image sequences. In this context, the term Human Sequence Evaluation (HSE) denotes those transformation processes involved in the textual descriptions of human behaviour from pixel values. Mainly, three co‐operating goals are involved in HSE: (i) a geometric description of the recorded human movements is first obtained; (ii) subsequently, these quantitative parameters are used to instantiate logic predicates; and (iii) textual representations are built upon these conceptual primitives for Natural‐Language (NL) text generation.
Local
Coimbra
Data
10 October 2008
Website
http://recpad2008.isr.uc.pt/
Logo
Could not load cover

RecPad 2002

12th Portuguese Conference on Pattern Recognition

RecPad 2000

11th Portuguese Conference on Pattern Recognition
recpad2000
Local
Porto
Data
5 May 2000