June 21, 2022 The COSPAL Project 1 Computer Vision Laboratory The COSPAL Project Michael Felsberg Computer Vision Laboratory Linköping University SWEDEN
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
April 21, 2023 The COSPAL Project 1
Computer Vision Laboratory
The COSPAL Project
Michael FelsbergComputer Vision Laboratory
Linköping UniversitySWEDEN
April 21, 2023 The COSPAL Project 2
Computer Vision Laboratory
Fact Sheet
• COSPAL acronym:COgnitiveSystems using Perception-Action Learning
• STREP IST-2003-004176
• Project Start:1. July 2004
April 21, 2023 The COSPAL Project 3
Computer Vision Laboratory
ConsortiumLinköping University, SwedenComputer Vision Laboratory (coordinator)Gösta GranlundChristian-Albrechts-University of Kiel, GermanyCognitive SystemsGerald SommerUniversity of Surrey, UKCentre for Vision, Speech, & Signal ProcessingVaclav HlavacCzech Technical University in Prague, Czech RepublicCenter for Machine PerceptionJosef Kittler
April 21, 2023 The COSPAL Project 4
Computer Vision Laboratory
Consortium
April 21, 2023 The COSPAL Project 5
Computer Vision Laboratory
Objectives
• Architectures for autonomous robots with cognitive capabilities
• Vision based systems
• Stationary manipulator
QuickTime™ and aYUV420 codec decompressor
are needed to see this picture.
April 21, 2023 The COSPAL Project 6
Computer Vision Laboratory
Feature input Output
Scenedescription
Actiongeneration
Approach
• Action precedes perception • Incremental learning and exploration• Bidirectional interface to reasoning unit
April 21, 2023 The COSPAL Project 7
Computer Vision Laboratory
Early Demonstrator
• Parallel progress in WPs
• Concentrate on architectural aspect
• Fully automatic learning and evaluation
QuickTime™ and a decompressor
are needed to see this picture.
April 21, 2023 The COSPAL Project 8
Computer Vision Laboratory
Final Demonstrator I
• Embedding into ‘real’ world
• Scalability of the architecture
• Stability and robustness
April 21, 2023 The COSPAL Project 9
Computer Vision Laboratory
QuickTime™ and aYUV420 codec decompressor
are needed to see this picture.
QuickTime™ and aYUV420 codec decompressor
are needed to see this picture.
Final Demonstrator II
QuickTime™ and aYUV420 codec decompressor
are needed to see this picture.
April 21, 2023 The COSPAL Project 10
Computer Vision Laboratory
Just a toy example?
Fact: this scenario can ‘easily’ be solvedby a traditional system design.
• Is it robust?
• Can it generalize?
• Is it scalable?
April 21, 2023 The COSPAL Project 11
Computer Vision Laboratory
Robustness/Generalization
?
April 21, 2023 The COSPAL Project 12
Computer Vision Laboratory
Challenge of COSPAL
… is NOT to solve the puzzle problem
… is NOT to do model registration and fitting
… but to develop a system architecture whichis able to semi-autonomously,step-by-step build up knowledgeand solution strategies
April 21, 2023 The COSPAL Project 13
Computer Vision Laboratory
Architecture
• Homogeneous network structures• Localized representations (channels)• Associations between channels to be learned
April 21, 2023 The COSPAL Project 15
Computer Vision Laboratory
Hardware
• PCs• Firewire cameras• RX90• 3D turntable
April 21, 2023 The COSPAL Project 16
Computer Vision Laboratory
Software & Interfaces
• Heterogeneous code modules:– C++– Matlab– Python
• Loosely coupled processes– Controlling process– XML files– Exception: servoing (PVM)
April 21, 2023 The COSPAL Project 17
Computer Vision Laboratory
Summary
• Autonomous cognitive system– Action-precedes-perception learning– Bidirectional interface to symbolic unit– Incremental learning at all levels
• Demonstrator as proof of architecture
• Architectural building blocks for systems in more serious applications
Related Documents
