Rochester Institute of Technology
Goal Directed Visual Search Based on Color Cues: Co-operative Effectes of Top-Down & Bottom-Up Visual Attention
Abstract
dc:description.abstractFocus of Attention plays an important role in perception of the visual environment. Certain objects stand out in the scene irrespective of observers' goals. This form of attention capture, in which stimulus feature saliency captures our attention, is of a bottom-up nature. Often prior knowledge about objects and scenes can influence our attention. This form of attention capture, which is influenced by higher level knowledge about the objects, is called top-down attention. Top-down attention acts as a feedback mechanism for the feed-forward bottom-up attention. Visual search is a result of a combined effort of the top-down (cognitive cue) system and bottom-up (low level feature saliency) system. In my thesis I investigate the process of goal directed visual search based on color cue, which is a process of searching for objects of a certain color. The computational model generates saliency maps that predict the locations of interest during a visual search. Comparison between the model-generated saliency maps and the results of psychophysical human eye -tracking experiments was conducted. The analysis provides a measure of how well the human eye movements correspond with the predicted locations of the saliency maps. Eye tracking equipment in the Visual Perceptual Laboratory in the Center for Imaging Science was used to conduct the experiments.
Degree
thesis:*- Name thesis:degree_name
- Computer Science (MS)
- Level thesis:degree_level
- Thesis
- Discipline thesis:degree_discipline
- Computer Science (GCCIS)
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Vaingankar, Vishal S
- Contributors dc:contributor
-
- Roger Gaborski
Identifiers
dc:identifier.*- Repository record dc:identifier
- https://repository.rit.edu/theses/7570