Calden presented “Psychophysical Evaluation of Saliency Algorithms” at VSS 2016
Authors: Calden Wloka, Sang-Ah Yoo, Rakesh Sengupta, Toni Kunic, and John K. Tsotsos
Title: Psychophysical Evaluation of Saliency Algorithms
Abstract: A great deal of effort has been spent evaluating the performance of saliency algorithms at predicting human fixations in natural images. However, many other aspects of human visual attention have received relatively little focus in the saliency literature but have been richly characterized by psychophysical investigations. Bruce et al.  have recommended the development of an axiomatic set of model constraints grounded in this body of psychophysical knowledge. We aim to provide a step towards this goal by linking human visual search response time to saliency algorithm output. Duncan and Humphreys  theorized that subject response time in visual search tasks is correlated with similarity between search items (with search time increasing both as targets become more similar to distractors and the heterogeneity of distractors increases). This result fits well with the widely held notion in the saliency model literature that saliency is largely driven by stimulus uniqueness, but has not been explicitly tested against the performance of saliency algorithms. To do so systematically, we need a well-characterized human performance curve for a given set of visual search stimuli.
Wolfe  provides a list of features which can, given su‑cient target-distractor differences, elicit effi‑cient search for singleton targets. These features therefore provide a strong candidate set upon which to test saliency algorithm performance. Arun  produced a well-characterized performance curve for the first such feature, orientation, by testing humans over a broad range of target-distractor orientation differences ranging from 7-60 degrees. Here we replicate Arun’s experiment, showing that saliency algorithm performance falls into three broad categories: those which cannot consistently find the target, those which consistently find the target but have no differentiated performance with target-distractor difference, and those which are able to fit a human performance-like curve. We can use these results to guide future saliency model development.
 S. P. Arun. Turning visual search time on its head. Vision Research, 74:86 92, 2012.
 Neil D. B. Bruce, Calden Wloka, Nick Frosst, Shan Rahman, and John K. Tsotsos. On computational modeling of visual saliency: Examining what’s right, and what’s left. Vision Research, 116:95 112, 2015.
 John Duncan and Glyn W Humphreys. Visual search and stimulus similarity. Psychological review, 96(3):433, 1989.
 Jeremy M. Wolfe. Visual search. In Harold Pashler, editor, Attention. Psychology Press, 1998.