Saliency Detection: A Boolean Map Approach

By Jianming Zhang, Stan Sclaroff. IEEE International Conference on Computer Vision, 2013

524 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition, Sensory Systems, Human-Computer Interaction

View Paper: https://www.semanticscholar.org/paper/75ee5cd7712b6ad13d64c3b4de24aa592a869fc1

How researchers see this paper

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

In Computer Vision and Pattern Recognition

In Sensory Systems

In Human-Computer Interaction

Citations in the corpus

Deep Gaze I: Boosting Saliency Prediction with Feature Maps Trained on ImageNet (2014, methodology): “The three top performing models after Deep Gaze I are in order of decreasing performance: BMS (82.57%, Zhang and Sclaroﬀ, 2013), Mixture of Saliency Models (82.09%, Han and Satoh, 2014), and eDN (81.92%, Vig et al., 2014).”
A deep multi-level network for saliency prediction (2016, background): “Some of these metrics compare the predicted saliency map with the ground truth map generated from [26] 0.4268 0.6935 0.7899 Baseline: GBVS [27] 0.4212 0.6303 0.7899 Baseline: Itti [2] 0.2046 0.6101 0.6669 ﬁxation points, while other directly compare the predicted saliency map with ﬁxation points…”
SalGAN: Visual Saliency Prediction with Generative Adversarial Networks (2017, methodology): “When considering AUC-Judd as the reference metric, results as of September 2016 on the MIT300 benchmark include only one non-deep learning technique (Boolean map saliency [31]) among the top-ten.”
Two-Stage Learning to Predict Human Eye Fixations via SDAEs (2016, methodology): “In this paper, all image pixels are represented in Commission Internationale d’Eclairage Lab color space, which is known for its perceptual uniformity [54].”
Shallow and Deep Convolutional Networks for Saliency Prediction (2016, unknown): “onvnet 0.5198 0.5957 0.6698 0.8291 0.8364 WHU IIP 0:4908 0:4569 0:6064 0:7759 0:7923 Rare 2012 Improved [30] 0:5017 0:5108 0:6644 0:8047 0:8148 Xidian 0:4617 0:4811 0:6809 0:7990 0:8051 Baseline: BMS [45] 0:4542 0:4268 0:6935 0:7699 0:7899 Baseline: GBVS [14] 0:4460 0:4212 0:6303 0:7816 0:7899 Basel”
SALICON: Saliency in Context (2015, background): “0 0.5 1 1.5 2 2.5 3 0.5 0.55 0.6 0.65 0.7 0.75 0.8 0.85 0.9 Blur Width (DVA) s A U C EYE MOUSE MOUSE−AMT BMS AWS AIM SIG SUN GBVS ITTI Figure 6.”
Attentive Systems: A Survey (2017, background): “…l–t, salient object detection methods). b Attention based on information maximization (AIM, Bruce et al. 2005), c boolean map based saliency (BMS, Zhang and Sclaroff 2013), d saliency based on region covariance (COV, Erdem and Erdem 2013), e graph based saliency (GB, Harel et al. 2006), f…”
Paying More Attention to Saliency: Image Captioning with Saliency and Context Attention (2017, background): “In addition to the exploitation of low-level features, several saliency models have also incorporated high-level concepts such as faces, people, and text [3, 21, 53].”
Predicting Human Eye Fixations via an LSTM-Based Saliency Attentive Model (2016, background): “Traditional saliency prediction methods have followed biological evidences by defining features that capture low-level cues such as color, contrast and texture or semantic concepts such as faces, people and text [8], [9], [10], [11].”
Surface defect saliency of magnetic tile (2018, methodology): “And BMS [21] is used to detect saliency objects by simulating human visual-attention system.”
SALICON: Reducing the Semantic Gap in Saliency Prediction by Adapting Deep Neural Networks (2015, methodology): “We compare our results with 14 saliency prediction models, namely BMS [42], AWS [11], AIM [2], RARE [30], LG [1], eDN [39], Judd [21], SigSal [15], CAS [12], CovSal [8], ΔQDCT [32], SUN [43], GBVS [14], and ITTI [16].”
Saliency detection by multi-context deep learning (2015, background): “A large number of approaches [63, 52, 40, 39, 32, 35, 60, 57, 56, 47, 41, 31, 27, 25, 24, 23, 11, 44, 17, 8, 13, 1, 21] are proposed to capture different saliency cues.”
Deep Level Sets for Salient Object Detection (2017, methodology): “Assuming image boundaries belong to the background, Zhang et al [60] and Tu et al [48] compute the difference between pixels and image boundaries to be saliency and achieve realtime testing speed.”
Salient Object Detection: A Benchmark (2015, methodology): “PR and ROC curves for BMS [100] and GB [93] over ECSSD.”
Deep Visual Attention Prediction (2017, methodology): “…ﬁxations, we evaluated it by comparison to 13 state-of-the-art models, including six classical models: ITTI [28], GBVS [38], Judd Model [24], BMS [57], CAS [41], AIM [56], and seven deep learning based models: DeeFix [44], SALICON [45], Mr-CNN [1], SalNet [15], Deep Gaze I [42], eDN [13], and…”
New Contour Cue-Based Hybrid Sparse Learning for Salient Object Detection (2019, methodology): “Boolean map saliency proposed by Zhang and Sclaroff [10] combines the Gestalt and ﬁgure-ground theories to form a uniﬁed model for both ﬁxation prediction and salient object detection.”
DeepFix: A Fully Convolutional Neural Network for Predicting Human Eye Fixations (2015, methodology): “Linear Correlation Coefﬁcient(CC) : The correlation coefﬁcient metric between G and S is given by : Image GBVS[4] eDN[33] BMS[49] Mr-CNN[34] DeepFix(proposed) Ground-truth 1) 2) 3) It gives a measure of the linear relationship between the two maps.”
Minimum Barrier Salient Object Detection at 80 FPS (2015, background): “Salient regions can be inferred by their dissimilarity in appearance [19, 16, 23], or their connectivity [39, 33, 36] with respect to the image boundary regions.”