de
en
Schliessen
Detailsuche
Bibliotheken
Projekt
Impressum
Datenschutz
zum Inhalt
Detailsuche
Schnellsuche:
OK
Ergebnisliste
Titel
Titel
Inhalt
Inhalt
Seite
Seite
Im Dokument suchen
Wischnewski, Marco: Where to look next? : Proto-object based priority in a TVA-based model of visual attention. 2011
Inhalt
Introduction
Scope and contributions of the thesis
Thesis outline
Cognitive neuroscience of visual attention and eye-movement control
Introduction
Spatial inhomogeneous visual processing
Overt attention
Saliency vs Priority
Low-level features vs object-based features
From low-level features to proto-objects
Visual search: where to look next?
Summary
A novel computational model of visual attention
Introduction
Spatial inhomogeneous processing
Proto-objects
Task-dependency by means of TVA and learning of object representations
The model's global architecture
Summary
The spatial inhomogeneous low-level feature map
Introduction
The spatial inhomogeneous pixel grid
Color and intensity
Summary
Proto-object segmentation by clustering
Introduction
A clustering algorithm for spatial inhomogeneously arranged data
The Gaussian pyramid
The computation of confidence values
Homogeneous regions by label propagation
Merging of regions
Filtering of regions
Parameters, variation, and robustness
The ``Global Effect''
Summary
Computation of mid-level features
Introduction
Weighted arithmetic mean by means of scaling
The mid-level features
Color and intensity
Size
Orientation
Shape
Summary
Learning the mid-level feature representations of natural objects
Introduction
A neural network approach for classification
The training stage
Summary
Object-based priority by means of TVA
Introduction
The modified TVA weight equation
The proximity effect: eccentricity-dependent weight modification
Summary
The priority-driven saccade
Introduction
Merging of proto-objects
The identity value
Overlapping of proto-objects
A new level of proto-objects
The attention priority map (APM)
Inhibition of return (IOR)
Winner-takes-all (WTA)
The landing position
Summary
The model performance
Introduction
Target-distractor discriminability
Performance examples
Summary
Summary and outlook
Notation
Image Library