Learning where to attend with deep architectures for image tracking

DOI10.1162/NECO_A_00312MaRDI QIDQ2919435zbMATH OpenOpenAlexDBLPWikidataFDO

Authors Misha Denil, Loris Bazzani, Hugo Larochelle, Nando De Freitas

Publication date 2 October 2012

Published in Neural Computation (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/1109.3737

Mathematics Subject Classification ID

Mathematical biology in general (92B99) Image processing (compression, reconstruction, etc.) in information and communication theory (94A08)

Abstract: We discuss an attentional model for simultaneous object tracking and recognition that is driven by gaze data. Motivated by theories of perception, the model consists of two interacting pathways: identity and control, intended to mirror the what and where pathways in neuroscience models. The identity pathway models object appearance and performs classification using deep (factored)-Restricted Boltzmann Machines. At each point in time the observations consist of foveated images, with decaying resolution toward the periphery of the gaze. The control pathway models the location, orientation, scale and speed of the attended object. The posterior distribution of these states is estimated with particle filtering. Deeper in the control pathway, we encounter an attentional mechanism that learns to select gazes so as to minimize tracking uncertainty. Unlike in our previous work, we introduce gaze selection strategies which operate in the presence of partial information and on a continuous action space. We show that a straightforward extension of the existing approach to the partial information setting results in poor performance, and we propose an alternative method based on modeling the reward surface as a Gaussian Process. This approach gives good performance in the presence of partial information and allows us to expand the action space from a small, discrete set of fixation points to a continuous domain.

Recommendations

• scientific article; zbMATH DE number 1977164
• Object-based visual attention for computer vision
• Learning to track the visual motion of contours
• Interpreting a dynamic and uncertain world: task-based control
• scientific article; zbMATH DE number 1977162

Cites work

• A decision-theoretic generalization of on-line learning and an application to boosting
• Gaussian processes for machine learning.
• scientific article; zbMATH DE number 1306865 (Why is no real title available?)
• Lipschitzian optimization without the Lipschitz constant
• Planning and acting in partially observable stochastic domains
• Reducing the Dimensionality of Data with Neural Networks
• The Nonstochastic Multiarmed Bandit Problem
• Training Products of Experts by Minimizing Contrastive Divergence

Cited in

(2)

• Neural bag-of-features learning
• Research and realization of video target detection system based on deep learning

This page was built for publication: Learning where to attend with deep architectures for image tracking

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2919435)