It is our interest to employ multiple visual features under a robust tracking framework.
Thus, one potential implication for future studies in this area is the benefit of including a control group of healthy adults, to treat them as a "normative" group to verify that the employed visual and auditory stimuli are approximately matched in their physical and/or perceptual features.
Balance in composition is one of the agreed-upon principles of aesthetics and is commonly employed as a visual feature in many computational aesthetics studies.
In order to assess the selection of visual features we employed a signal detection framework based on the Receiver Operator Characteristic (ROC) (Le Meur, Le Callet & Barba, 2007).
This paper deals with the use of invariant visual features for visual servoing.
Afterwards, we employ a weighted strategy to integrate the visual features with multimodal features.
We employ an adaptive integration of the multiple visual features mentioned above, i.e., democratic integration.
We employ a number of low and high-level visual features such as static saliency, motion, faces, humans and text, some of which have not been previously used in dynamic saliency estimation.
The latent topics are learnt, based on the relationship between the images and words, and are employed to bridge the gap between low-level visual features and high-level semantics.
The geometric information and visual information are extracted by Convolutional Neural Networks (CNNs) and Convolutional Deep Belief Networks (CDBNs), respectively, and then two independent Deep Belief Networks (DBNs) are employed to learn high-level features from geometric and visual features.
It is widely accepted that the early mammalian visual system employs a series of neural circuits to extract elementary visual features, such as edges and motion [1, 2].
