petia radeva
play

Petia Radeva University of Barcelona & Computer Vision Center - PowerPoint PPT Presentation

Feb 17, 2023 •135 likes •488 views

Computer Vision in Barcelona Petia Radeva University of Barcelona & Computer Vision Center Presentation to: AComIn: Advanced Computing for Innovation The Computer Vision Center CVC is a legally independent, non-profit institution

  1. Computer Vision in Barcelona Petia Radeva University of Barcelona & Computer Vision Center Presentation to: AComIn: Advanced Computing for Innovation

  2. The Computer Vision Center CVC is a legally independent, non-profit � institution founded in 1995 by the Generalitat de Catalunya and the Universitat Autònoma de Barcelona (UAB). � Specialists and leaders in the field of computer vision. � More than 130 multidisciplinary researchers and technicians of different nationalities. � Advanced resources in Computer Vision hardware and software 2000m 2 devoted to R&D of excellence on � Computer Vision. Presentation to AComIn: Advanced Computing for Innovation

  3. Vision & Mission Vision The CVC claims to be recognized as: R+D European leader in Computer Vision � � A consolidated public institution guided by the quality and volume of its scientific production on computer vision; as well as by its knowledge transfer and its drive for society development � A reference in the exchange, integration and mobility of researchers on an international scale � An institution within the UAB framework that offers an innovative, attractive and applied postgraduate education with international vocation � A model that is driven by the clients satisfaction A link to the industrial network � Mission Carry out research of renowned quality and international impact . Transfer knowledge towards companies and society. Train high-level scientists Presentation to AComIn: Advanced Computing for Innovation

  4. The CVC in the UAB Research Park and the UAB30 initiative CVC Presentation to AComIn: Advanced Computing for Innovation

  5. Main Activities Cutting edge research Medical image analysis � Visual object recognition � Document analysis � Image understanding � Color and texture � � On board vision � Visual perception � Machine vision Multimedia indexing and retrieval � Video surveillance interpretation � Interactive 3D visualization and � augmented reality Technological Transfer & Consultancy Education & Training � Master Program (UAB & IIIA) � PhD Program (UAB) Presentation to AComIn: Advanced Computing for Innovation

  6. Organization Chart Steering Committee Horst Bunke Lluís Ferrer Lluís Torner Jan Oosterveld Walter Kropatsch Joan Majó Advisory Committee Scientific Committee Filiberto Pla Ramón Ollé Pere Brunet Antoni Subirà Director Andreas Dengel Executive Committee R&D Committee Deputy Director General Manager Research Office Administration ADAS RRHH Technology Transfer CiC Marketing & Comunication DAG Technical Supp. & Laboratory ISE Computer Services MiLab Maintenance MaVis OR IAM Presentation to AComIn: Advanced Computing for Innovation

  7. Presentation to AComIn: Advanced Computing Personnel for Innovation

  8. International impact People Research Institutions Companies Presentation to AComIn: Advanced Computing for Innovation

  9. CVC Features Specialization in Computer Vision as a horizontal field. � Focus to both Research and Development . � � Mixed (engage and associate) researchers and developers. Collaboration with Technological Companies . � A middle size and flexible Institution. � A reference centre for companies. � Multidisciplinary personnel: Computer Science, Mathematics, Physics, Telecom and � Electronic Engineering specialists. Presentation to AComIn: Advanced Computing for Innovation

  10. Models of R and D Without Cooperation New Marketing Ideas actions R D Companies Competitive Knowledge Know-How Funds Funds Research Innovative Activity Activity PUBLICATIONS PROTOTYPES Presentation to AComIn: Advanced Computing for Innovation

  11. CVC Model of R&D Cooperation New Marketing Ideas Actions R D CVC Competitive Private Funds Funds Knowledge/Know-How Research Development Activity Activity PUBLICATIONS PROTOTYPES Presentation to AComIn: Advanced Computing for Innovation

  12. Models of R&D Without Cooperation Presentation to AComIn: Advanced Computing for Innovation

  13. CVC: Model of R&D Presentation to AComIn: Advanced Computing for Innovation

  14. The Open Innovation Paradigm IP in-sourced for Products in-sourced development for scale up Research Innovation Commercialization in-sourced ideas IP licensing Technology and technologies spin-outs Knowledge Ideas Market Presentation to AComIn: Advanced Computing for Innovation

  15. Publications Journals National conferences International conferences Books Chapters Presentation to AComIn: Advanced Computing for Innovation

  16. Transfer indexes Presentation to AComIn: Advanced Computing for Innovation

  17. Current International Competitive Projects 7 FRAMEWORK PROGRAM (FP7-PEOPLE-2008-IAPP: 230653) “ADAO: Administrative Document Automate Optimization” 2009-2013 - 184k € ITEA2 European Project VICOMO (TSI-020400-2009-133) “ViCoMo: Visual Context Modeling” 2009-2011 - 160k € EUREKA European Project SCANPLAN (Eureka E!-4462) “ScanPlan: Architectural plans recognition” 2008-2011 – 220k € 7 FRAMEWORK PROGRAM (FP7-269796) “5CofM: Five Centuries of Marriages” 2011-2015 – 397k € ITEA2 European Project DICOMA (Itea2-10031) “DICOMA: Disaster Control Management” 2011 – 35k € Presentation to AComIn: Advanced Computing for Innovation

  18. Why CVC The main asset and guarantee of our work are the confidence placed by our partners for over 15 years, experiencing at first hand our expertise and professionality. � More than 300 projects and feasibility studies. 7 Spin-offs already launched. � � More than 150 companies among our customers. Presentation to AComIn: Advanced Computing for Innovation

  19. Spin-offs VISUAL TAGGING SERVICES (2012): Mobile apps. CLOUD SIZING SERVICES (2012): Sizing clothing. DAVANTIS (2005): Smart surveillance. INSPECTA (2003): Cork quality control. ICAR Vision Systems (2002): Systems for personal documents. VISUAL CENTURY (2001): Video indexing. VISIÓ I ROBÒTICA APLICADA (VyRA) (1998): Computer Vision solutions. Presentation to AComIn: Advanced Computing for Innovation

  20. Collaborations with the Catalan R&D system CERCA OTHERS CENTRES TECNIO BSC ASCAMM CED IIIA LEITAT CREAF CRIC BCNDigital I2CAT IRI AIICA CTM CTFC Fundació CIM CETEMMSA CTFC UDIAT CISTIB IGTP IMIM IR-Sant Pau VHIR TECNIO GROUPS HOSPITALS IRTA CEPHIS (CAIAC) TRIAS i PUJOL ViCOROB PARC TAULÍ CD6 SANT JOAN DE DEU UDTIA (IIIA) VALL D’HEBRON VERGE DE LA CINTA DE TOROSA Presentation to AComIn: Advanced Computing for Innovation

  21. Income Evolution (euros) Presentation to AComIn: Advanced Computing for Innovation

  22. Income Evolution (euros) Presentation to AComIn: Advanced Computing for Innovation

  23. Focalització sectorial dels grups d’R+D Sectorial focus of the R&D groups Mobility and ADAS : Advanced Production MV : Machine Vision Driver Assistance transportation Systems MV ADAS ISE : Image Sequence CIC : Colour In Context Evaluation Basic research ISE DAG : Document MILab : Medical Analysis Group Imaging Laboratory CiC DAG MILab VOR IAM : Interactive & VOR : Visual Object augmented modeling Recognition IAM Intelligent Health contents and media and wellfare Presentation to AComIn: Advanced Computing for Innovation

  24. Presentation to AComIn: Advanced Computing for Innovation

  25. Collaboration with CVC The CVC has the objective to contribute innovation to companies and society. Our ways to collaborate are: � Ad-hoc projects: Specific and customized projects. � Sponsored Research : Externalized R&D projects under open innovation paradigm. � Collaborative projects: under co-funding of public agencies. � Industrial studentship: Master internships, industrial doctorate. Presentation to AComIn: Advanced Computing for Innovation

  26. Computer Vision Challenges and Opportunities Presentation to AComIn: Advanced Computing for Innovation

  27. Medical imaging Presentation to AComIn: Advanced Computing for Innovation

  28. Security and Surveillance: Biometric descriptors Fingerprint identification, face recognition, iris and retinal scanning, hand geometrics, ear geometrics, signature recognition, voice identification, DNS identification, human scent, typing characteristics or gait recognition are not the privileges of science fiction any more. 3 2 1 0 -1 -2 -3 -2 -1.5 -1 -0.5 0 0.5 1 1.5 2 2.5 Presentation to AComIn: Advanced Computing for Innovation

  29. Security and Surveillance Face Screening US-VISIT Check point control 3M Border control Adaptive Recognition Hungary Viisage ( L-1 Identity Solutions) Swim Pool drown Tracking and Action Detection > “Alert!!! > A Person is forcing a Ticket Machine!!” Poseidon-Tech Presentation to AComIn: Advanced Computing for Innovation

  30. Media and Entertainment ARmusement Gaming interfaces Augmented Reality Sony EyeToy (for PlayStation). Allows players to interact with games using motion, color detection and sound. Digital Photography Photosynth (Microsoft LiveLabs) GestureFX. Gesture interfaces in ground, wall or tables. Presentation to AComIn: Advanced Computing for Innovation

  31. Media and Entertainment Augmented TV PVI: Virtual advertising in real tv images. Hawk Eye: ball tracking in sport games. Presentation to AComIn: Advanced Computing for Innovation

Recommend

Uncertainty-Aware Food Recognition by Deep Learning Petia Radeva, Collaboration with: Eduardo

Uncertainty-Aware Food Recognition by Deep Learning Petia Radeva, Collaboration with: Eduardo

Uncertainty-Aware Food Recognition by Deep Learning Petia Radeva, Collaboration with: Eduardo Aguilar, Marc Bolaos University of Barcelona & Computer Vision Center radevap@gmail.com 11:01 The Diabetes pandemy Diabetic people need to

907 views • 51 slides
Food Recognition using Fusion of Classifiers based on CNNs Eduardo Aguilar, Marc Bolaos, and

Food Recognition using Fusion of Classifiers based on CNNs Eduardo Aguilar, Marc Bolaos, and

ICIAP 2017 Food Recognition using Fusion of Classifiers based on CNNs Eduardo Aguilar, Marc Bolaos, and Petia Radeva Presented by: Petia Radeva Contents The problem of automatic food analysis Convolutional Neural Network (CNN)

815 views • 27 slides
Food Ingredients Recognition through Multi-label Learning Marc Bolaos, Aina Ferr and Petia

Food Ingredients Recognition through Multi-label Learning Marc Bolaos, Aina Ferr and Petia

Food Ingredients Recognition through Multi-label Learning Marc Bolaos, Aina Ferr and Petia Radeva Motivation Motivation Motivation INPUT OUTPUT Computer List of Nutritional Vision Ingredients Composition Algorithm dishes

721 views • 23 slides
purposes Pe$a Radeva University of Barcelona &

purposes Pe$a Radeva University of Barcelona &

Visual Lifelogging: using non-medical images for medical purposes Pe$a Radeva University of Barcelona & Computer Vision Center Medical

869 views • 47 slides
Pattern-based Analysis of the Control- flow Perspective of UML Activity Diagrams Petia Wohed

Pattern-based Analysis of the Control- flow Perspective of UML Activity Diagrams Petia Wohed

Pattern-based Analysis of the Control- flow Perspective of UML Activity Diagrams Petia Wohed UHP (SU/KTH) Wil M.P. van der Aalst TUE & QUT Marlon Dumas QUT Arthur H.M. ter Hofstede QUT Nick Russell QUT CRICOS No. 00213J Queensland

531 views • 15 slides
The Columbia-GWU System at the 2016 TAC KBP BeSt Evaluation Owen Rambow, Tao Yu, Axinia Radeva,

The Columbia-GWU System at the 2016 TAC KBP BeSt Evaluation Owen Rambow, Tao Yu, Axinia Radeva,

Data English Sentiment 1 English Sentiment 2 Chinese Sentiment Spanish Sentiment English Belief Chinese Belief Spanish Belief The Columbia-GWU System at the 2016 TAC KBP BeSt Evaluation Owen Rambow, Tao Yu, Axinia Radeva, Sardar Hamidian,

657 views • 31 slides
Meta-Modelling as a Means for Improved Communication and Interoperability The Case of Frisco

Meta-Modelling as a Means for Improved Communication and Interoperability The Case of Frisco

Meta-Modelling as a Means for Improved Communication and Interoperability The Case of Frisco Petia Wohed & Birger Andersson EMOI, 13 June 2005, Porto 1 Background of our work There is terminological fuzziness in IS engineering

290 views • 10 slides
Reducing Noise in Labels and Features for a Real World Dataset: Application of NLP Corpus

Reducing Noise in Labels and Features for a Real World Dataset: Application of NLP Corpus

Reducing Noise in Labels and Features for a Real World Dataset: Application of NLP Corpus Annotation Methods Rebecca J. Passonneau, Cynthia Rudin, Axinia Radeva, and Zhi An Liu Center for Computational Learning Systems (CCLS) Columbia University

381 views • 20 slides
Roman Gorbonosov on behalf of the Beams Department Controls Group Based on the input from

Roman Gorbonosov on behalf of the Beams Department Controls Group Based on the input from

Roman Gorbonosov on behalf of the Beams Department Controls Group Based on the input from M.Arruat, V.Baggiolini, JC.Bau, M.Buttner, P.Charrue, S.Deghaye, E.Hatziangeli, G.Kruk, M.Lamont, A.Radeva, U.Raich, C.Roderick, J.Serrano, W.Sliwinski,

919 views • 72 slides
Towards a Professional Development Framework for Leaders of Degree Programmes Susan Moron

Towards a Professional Development Framework for Leaders of Degree Programmes Susan Moron

Teaching and Learning Exchange Towards a Professional Development Framework for Leaders of Degree Programmes Susan Moron Garcia, University of Central Lancashire Petia Petrova, University of the West of England Elizabeth Staddon, University

338 views • 17 slides
Barriers to Household Risk Management: Evidence from India Shawn Cole Xavier Gine Jeremy Tobacman

Barriers to Household Risk Management: Evidence from India Shawn Cole Xavier Gine Jeremy Tobacman

Barriers to Household Risk Management: Evidence from India Shawn Cole Xavier Gine Jeremy Tobacman (HBS) (World Bank) (Wharton) Petia Topalova Robert Townsend James Vickery (IMF) (MIT) (NY Fed) Presentation by Xavier Gine Index Insurance 4

568 views • 39 slides

More recommend