Seite - 5 - in Proceedings - OAGM & ARW Joint Workshop 2016 on "Computer Vision and Robotics“
Bild der Seite - 5 -
Text der Seite - 5 -
Hierarchical
Compositional Representations of Structure for
Computer Vision and
Robotics
Ales
Leonardis
University of Birmingham, School of Computer Science, United Kingdom
a.leonardis@cs.bham.ac.uk
Abstract
Modelling, learning, recognising,
and categorising
visual
entities has been
an
area of intensive re
search in the vision and robotics communities for several decades. While successful partial
solutions tailored for particular tasks and specific scenarios have appeared in recent years, more
general solutions are yet to be developed. Ultimately, the goal is to design and implement proper
structures and mechanisms that would enable efficient learning, inference, and, when necessary,
augmentation
and
modifications of the acquired visual knowledge in general scenarios.
Recently, it
has become increasingly clear that possible solutions should be sought in the framework of
hierarchical architectures. Among various design choices related to hierarchies, compositional
hierarchies show a great promise in terms of scalability, real time performance, efficient
structured on line learning, shareability, and knowledge
transfer. In
this talk I will first present our
work on compositional hierarchies related to visual representations of 2D and 3D object shapes
for recognition and grasping and then conclude with some ideas towards generalising the
proposed approach to other visual entities
and modalities.
5
Proceedings
OAGM & ARW Joint Workshop 2016 on "Computer Vision and Robotics“
- Titel
- Proceedings
- Untertitel
- OAGM & ARW Joint Workshop 2016 on "Computer Vision and Robotics“
- Autoren
- Peter M. Roth
- Kurt Niel
- Verlag
- Verlag der Technischen Universität Graz
- Ort
- Wels
- Datum
- 2017
- Sprache
- englisch
- Lizenz
- CC BY 4.0
- ISBN
- 978-3-85125-527-0
- Abmessungen
- 21.0 x 29.7 cm
- Seiten
- 248
- Schlagwörter
- Tagungsband
- Kategorien
- International
- Tagungsbände