CIS 3365 Lecture Notes - Lecture 15: Gabor Filter, Visual Search, Frequency Response

• School: University of Houston
• Department: Computer Information Systems
• Course: CIS 3365
• Professor: Xuqing Wu

Document Summary

Certain visual features pop out (pre-attentive features) Data variables should (usually) be mapped to pre- attentive features (they are processed fast) Restrictions (if you want pre-attentive design): conjunction searches are usually not pre-attentive, one can effectively display only limited number of visual variables, with limited accuracy. Integral visual dimensions interfere with each other - you should use separable dimensions. Some visual objects are processed pre-attentively, before the conscious attention. Pre-attentive processing speed is independent of the number of distractors. Some limitations: conjunction search usually not possible, some channels interfere with each other, limited resolution on channels. Some (but not all) pre-attentive features can be. These arrays can be modeled by the gabor model. The receptive field widths are around 2 cycles. Uncertainty principle: for constant data density either orientation or size can be specified precisely, but not both. Ideal garbor model has 3 parameters: contrast (c), orientation (o), and size (s) (or frequency i/s).