Saliency, attention, and visual search: an information theoretic approach.

By Neil D. B. Bruce, John K. Tsotsos. Journal of Vision, 2009

944 citations recorded by the source corpus.

A proposal for saliency computation within the visual cortex is put forth based on the premise that localized saliency computation serves to maximize information sampled from one's environment. The model is built entirely on computational constraints but nevertheless results in an architecture with cells and connectivity reminiscent of that appearing in the visual cortex. It is demonstrated that a variety of visual search behaviors appear as emergent properties of the model and therefore basic principles of coding and information transmission. Experimental results demonstrate greater efficacy in predicting fixation patterns across two different data sets as compared with competing models.

Subfields: Cognitive Neuroscience, Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/25e67f4ca6f45533cbe5cec4c6d169b7daf3af2b

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Cognitive Neuroscience

In Computer Vision and Pattern Recognition

Citations in the corpus

Two-Stage Learning to Predict Human Eye Fixations via SDAEs (2016, methodology): “Hou and Zhang [19] and Bruce and Tsotsos [37] measured the global contrast based on the information theoretic model which regards the computation of saliency as the process of maximizing information sampled from one’s environment.”
Fast Binary Coding for the Scene Classification of High-Resolution Remote Sensing Imagery (2016, methodology): “…method (context-aware) is evaluated, we investigate ﬁfteen kinds of typical methods, which are AC [53], attention by information maximization (AIM) [54], context-aware (CA) [52], context-based (CB) [55], discriminative regional feature integration (DRFI) [56], frequency-tuned (FT) [57],…”
Active inference and epistemic value (2015, background): “An alternative perspective on this epistemic quantity is afforded by Bayesian surprise; namely, the KL divergence between prior and posterior beliefs (Bruce & Tsotsos, 2009; Itti & Baldi, 2009).”
“Important stuff, everywhere!” Activity recognition with salient proto-objects as context (2014, background): “Since visual attention and object recognition are tightly linked processes in the human visual system, there is an increasing interest in integrating both concepts to increase the performance of computer vision systems.”
Deep Visual Attention Prediction (2017, methodology): “…400 Yes Yes SalNet [15] 320 × 240 Yes Yes Deep Gaze I [42] full size Yes Yes Saliency Uniﬁed [16] 417 × 417 Yes Yes eDN [13] 512 × 384 Yes Yes AIM [56] full size Yes No Judd Model [24] 200 × 200 Yes No BMS [57] w = 600 No No CAS [41] max { w,h } =250 No No GBVS [38] full size No No ITTI [28] full…”
Global Contrast Based Salient Region Detection (2015, methodology): “In this work, we extensively evaluated our saliency detection method on three different types of benchmark datasets, and compared it against 15 alternate methods — SR [30], IT [17], IM [23], SUN [27], AC [31], SeR [26], AIM [22], GB [29], LC [34], CA [32], FT [33], SWD [28], SEG [25], MSS [24], LP [”
Methods for comparing scanpaths and saliency maps: strengths and weaknesses (2012, methodology): “We compare the performance of four state-of-the-art models: Itti ’ s model (Itti, Koch, & Niebur, 1998), Le Meur ’ s model (Le Meur et al., 2006), Bruce ’ s model (Bruce & Tsotsos, 2009), and Judd ’ s model (Judd et al., 2009).”
Attention, Uncertainty, and Free-Energy (2010, background): “This is (heuristically) related to another measure of saliency, proposed by Bruce and Tsotsos (2009), who suggest that visual searches are attracted to areas of the visual field which maximize the information sampled.”
Image Signature: Highlighting Sparse Salient Regions (2012, methodology): “We compare our saliency maps generated from the image signature to the following published saliency algorithms: the original Itti-Koch saliency model [8] (denoted Itti), Dynamic Visual Attention model [15] (denoted DVA), Graph-Based visual saliency [16] (denoted GBVS), Attention based on Information”
Saliency Detection: A Boolean Map Approach (2013, methodology): “The negative logarithm of the probability, known as Shannon's selfinformation , is used to measure the improbability of a local patch as a bottom-up saliency cue in [6] and [39].”
Center-surround divergence of feature statistics for salient object detection (2011, methodology): “Bruce and Tsotsos have presented an interesting approach that computes the self-information of image regions with respect to their surround [4].”
Deep Gaze I: Boosting Saliency Prediction with Feature Maps Trained on ImageNet (2014, neutral): “(67.90%, Garcia-Diaz et al., 2012), RARE2012 (66.54%, Riche et al., 2013), and AIM (65.64%, Bruce and Tsotsos, 2009).”