Simulation of Human Visual Search in Cluttered Backgrounds

Abstract

Work in progress at Georgia Tech to develop a model of human pattern perception, visual search, and detection is reviewed. The model's algorithms are based on research on low-level visual processes. Recent advances in that field have led to the development of computational models of the image processing performed by the visual system from the cornea to the striate cortex. The model also incorporates recent advances from research on visual search. The organization of the model is described, and the results of some preliminary tests are presented.

Visual Attention to Plain and Ornamented Human Bodies: An Eye-Tracking Study

“…Although this might be related to the phenomenon that attention tends to jump to something new (von Helmholtz, 1925) or something which differs from a plain surface (Doll, Worther, & Schmieder, 1993), the constant and systematic variation supports the basic hypothesis, that specifically tattoos draw attention, since tattoos were responsible for the significant main effect. In particular, participants devoted significantly more attention to figures with tattoos than those with accessories, scars, and plain body images.-plain bodies received the least attention, followed by scarred, accessorized, and tattooed with the longest Gaze Duration.…”

“…In particular, participants devoted significantly more attention to figures with tattoos than those with accessories, scars, and plain body images.-plain bodies received the least attention, followed by scarred, accessorized, and tattooed with the longest Gaze Duration. Although this might be related to the phenomenon that attention tends to jump to something new (von Helmholtz, 1925) or something which differs from a plain surface (Doll, Worther, & Schmieder, 1993), the constant and systematic variation supports the basic hypothesis, that specifically tattoos draw attention, since tattoos were responsible for the significant main effect. The ranking of conditions suggests that 'natural' alterations might bear less differentiated information and so do not attract as much attention.…”

“…This holds true most of the time, but not always (Deubel & Schneider, 1996). Also, features that simply deviate from plain surfaces are more likely to be fixated (Doll, et al, 1993). Furthermore, the context of a scene (Rayner, 1998) and the complexity of an object might alter fixation times, with more detailed objects which are harder to integrate into a scene demanding longer fixation for identification than simple objects with fewer visual elements (Solso, 1999).…”

“…Furthermore, the context of a scene (Rayner, 1998) and the complexity of an object might alter fixation times, with more detailed objects which are harder to integrate into a scene demanding longer fixation for identification than simple objects with fewer visual elements (Solso, 1999). Also, features that simply deviate from plain surfaces are more likely to be fixated (Doll, et al, 1993). Familiarity with the object might influence fixation number and time (Greene & Rayner, 2001).…”

A breadth-first survey of eye-tracking applications

“…However, there is doubt in the literature as to whether human visual search can be described as an integration of independently processed features (Van Orden & DiVita, 1993). Visual search, even over natural stimuli, is at least partially deterministic, rather than completely random (Doll, Whorter, & Schmieder, 1993). Determinism stems from either or both of two kinds: The observer's strategy determines the search pattern (as in reading), and /or the direction of the next saccade is based on information gained through peripheral vision about surrounding stimuli.…”