Visual Attention Through Uncertainty Minimization in Recurrent Generative Models

@article{Standvoss2020VisualAT,
  title={Visual Attention Through Uncertainty Minimization in Recurrent Generative Models},
  author={K. Standvoss and Silvan C. Quax and M. V. van Gerven},
  journal={bioRxiv},
  year={2020}
}
Allocating visual attention through saccadic eye movements is a key ability of intelligent agents. Attention is both influenced through bottom-up stimulus properties as well as top-down task demands. The interaction of these two attention mechanisms is not yet fully understood. A parsimonious reconciliation posits that both processes serve the minimization of predictive uncertainty. We propose a recurrent generative neural network model that predicts a visual scene based on foveated glimpses… Expand
A Review of Uncertainty Quantification in Deep Learning: Techniques, Applications and Challenges
TLDR
This study reviews recent advances in UQ methods used in deep learning and investigates the application of these methods in reinforcement learning (RL), and outlines a few important applications of UZ methods. Expand

References

SHOWING 1-10 OF 76 REFERENCES
Computational modelling of visual attention
TLDR
Five important trends have emerged from recent work on computational models of focal visual attention that emphasize the bottom-up, image-based control of attentional deployment, providing a framework for a computational and neurobiological understanding of visual attention. Expand
Top-down control of visual attention in object detection
TLDR
The results validate the proposition that top-down information from visual context modulates the saliency of image regions during the task of object detection. Expand
Saccadic Predictive Vision Model with a Fovea
TLDR
Comparisons of the Error Saccade Model on Predictive Vision Models with and without a fovea show that a fvea-like structure in the input level of the PVM improves the error model's ability to pursue detailed objects in its view. Expand
Objects predict fixations better than early saliency.
TLDR
The eye position of human observers is measured while they inspect photographs of common natural scenes to suggest that early saliency has only an indirect effect on attention, acting through recognized objects. Expand
Where to look next? Eye movements reduce local uncertainty.
TLDR
A rigorous analysis of sequential fixation placement reveals that observers may be using a local rule: fixate only the most informative locations, that is, reduce local uncertainty. Expand
Top-down influences on visual processing
TLDR
The various top-down influences exerted on the visual cortical pathways are discussed and the dynamic nature of the receptive field is highlighted, which allows neurons to carry information that is relevant to the current perceptual demands. Expand
Attention, Uncertainty, and Free-Energy
TLDR
It is shown that if the precision depends on the states, one can explain many aspects of attention, including attentional bias or gating, competition for attentional resources, attentional capture and associated speed-accuracy trade-offs. Expand
Gaze Control as Prediction
  • J. Henderson
  • Medicine, Psychology
  • Trends in Cognitive Sciences
  • 2017
TLDR
This Opinion article outlines an alternative approach proposing that gaze control in natural scenes can be characterized as the result of knowledge-driven prediction and provides a theoretical framework for bridging gaze control and other related areas of perception and cognition at both computational and neurobiological levels of analysis. Expand
Shifts in selective visual attention: towards the underlying neural circuitry.
TLDR
This study addresses the question of how simple networks of neuron-like elements can account for a variety of phenomena associated with this shift of selective visual attention and suggests a possible role for the extensive back-projection from the visual cortex to the LGN. Expand
Deep-BCN: Deep Networks Meet Biased Competition to Create a Brain-Inspired Model of Attention Control
  • Hossein Adeli, G. Zelinsky
  • Computer Science
  • 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW)
  • 2018
TLDR
With Deep-BCN a DNN implementation of BCT now exists, which can be used to predict the neural and behavioral responses of an attention control mechanism as it mediates a goal-directed behavior-in this study the eye movements made in search of a target goal. Expand
...
1
2
3
4
5
...