Search results for "Saliency maps"
showing 3 items of 3 documents
Fully automatic saliency-based subjects extraction in digital images
2013
In this paper we present a novel saliency-based technique for the automatic extraction of relevant subjects in digital images. We use enhanced saliency maps to determine the most relevant parts of the images and an image cropping technique on the map itself to extract one or more relevant subjects. The contribution of the paper is two-fold as we propose a technique to enhance the standard GBVS saliency map and a technique to extract the most salient parts of the image. The GBVS saliency map is enhanced by applying three filters particularly designed to optimize the performance for the task of relevant subjects extraction. The extraction of relevant subjects is demonstrated on a manually ann…
The colour of light for urban spaces
2013
Light in urban spaces was first introduced to prevent crimes against both people and properties and, while there is no study that confirms crimes' reduction in lit spaces, it is undeniable that people feel safer in lit urban spaces. Over the years it became evident that light can also improve the liveability of cities and enhance their characteristics, therefore in the last years a growing attention was paid to urban lighting design taking into account the environmental impact and energy saving requirements too. The will to use the lighting design project to improve the liveability of historical centres it’s clear: the city during the night is no more conceived like a sum of lit architectur…
Saliency Map for Visual Perception
2015
Human and other primates move their eyes to select visual information from the scene, psycho-visual experiments (Constantinidis, 2005) suggest that attention is directed to visually salient locations in the image. This allows human beings to bring the fovea onto the relevant parts of the image, to interpret complex scenes in real time. In visual perception, an important result was the discovery of a limited set of visual properties (called pre attentive), detected in the first 200-300 milliseconds of observation of a scene, by the low-level visual system. In last decades many progresses have been made into research of visual perception by analyzing both bottom up (stimulus driven) and top d…