Predicting human gaze using low-level saliency combined with face detection

By Moran Cerf, Jonathan Harel, Wolfgang Einhäuser, Christof Koch. Neural Information Processing Systems, 2007

557 citations recorded by the source corpus.

View Paper: https://www.semanticscholar.org/paper/a48dc9c898d6ff1cdb96dcf8a297bafb8ac1b9e2

How researchers see this paper

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

Citations in the corpus

A Benchmark of Computational Models of Saliency to Predict Human Fixations (2012, background): “[11][9][12][10] confirmed that faces and text strongly attract attention and showed that they were difficult to ignore even when doing so imposes a cost.”
Exploiting local and global patch rarities for saliency detection (2012, background): “, faces, text [45], people, and cars [3], affective and emotional stimuli or actions within scenes [6]) when freely viewing scenes.”
State-of-the-Art in Visual Attention Modeling (2013, methodology): “Some studies have also added specific features for directing attention like skin hue [120], face [167], horizontal line [93], wavelet [133], gist [92], [93], center-bias [123], curvature [124], spatial resolution [125], optical flow [15], [126], flicker [119], multiple superimposed orientations (cro”
Learning to predict where humans look (2009, methodology): “Cerf et al. [3] improve upon the Itti model by adding face detection to the model.”
SALICON: Reducing the Semantic Gap in Saliency Prediction by Adapting Deep Neural Networks (2015, methodology): “In fact, object detectors have been shown to play an important role in improving saliency prediction [40], and several computational models have successfully incorporated object detectors into saliency models [4, 21, 44].”
Saliency Detection: A Boolean Map Approach (2013, methodology): “Although several compared methods have similar performance as BMS on some of the datasets, e.g. AWS [10] on MIT and Kootstra, ΔQDCT [33] on Toronto and Cerf, GBVS [13] on ImgSal, none of them achieves top performance on all five datasets.”
Revisiting Video Saliency Prediction in the Deep Learning Era (2021, background): “Further, human beings are important high-level stimuli [90], [91] in scene free-viewing.”
Saliency Detection by Multitask Sparsity Pursuit (2012, methodology): “Since most images of the FiFa data set contain faces, which is similar in [34], we use a face detector to get a face mask for each image as top-down priors.”
Visual Saliency Based on Scale-Space Analysis in the Frequency Domain (2013, unknown): “F”
Predicting Human Eye Fixations via an LSTM-Based Saliency Attentive Model (2016, methodology): “Other methods enriched predictions exploiting semantic classifiers for detecting higher level concepts such as faces, persons, cars and horizons [33], [10], [34], [9], [35].”
The Secrets of Salient Object Segmentation (2014, methodology): “In this section, we benchmark 7 fixations algorithms, AWS [13], AIM [6], SIG [15], DVA [16], GBVS [14], SUN [26], and ITTI [17] on 5 datasets, Bruce [6], Cerf [8], IS [20], Judd [18], and our PASCAL-S.”
Revisiting Video Saliency: A Large-Scale Benchmark and a New Model (2018, background): “Further, human beings are important high-level stimuli [10, 8] in free-viewing.”
Get real! Resolving the debate about equivalent social stimuli (2009, background): “Based on previous work that has shown that people have a profound bias to look at eyes and heads regardless of task instruction (Birmingham et al., 2008b; Cerf et al., 2008), it is possible that any selection bias for arrows may be fleeting.”
Action from Still Image Dataset and Inverse Optimal Control to Learn Task Specific Visual Scanpaths (2013, background): “Most of the image datasets available have been collected under free-viewing, and the few task controlled ones[14, 7] have been designed for small scale studies.”
Salient object detection: A survey (2014, background): “[159] show that faces and text attract human gaze.”
Gaze allocation in natural stimuli: Comparing free exploration to head-fixed viewing conditions (2009, methodology): “As example for a bottom-up model, we tested Itti and Koch's ( 2000 ) saliency map, using its original version rather than its more recent developments (Cerf, Harel, Einhauser, & Koch, 2008 ; Itti, 2005 ; Itti & Baldi, 2006 ).”