Graph-Based Visual Saliency

By Jonathan Harel, Christof Koch, Pietro Perona. 2006

3,868 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition

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

Citations in the corpus

ImageNet Large Scale Visual Recognition Challenge (2014, background): “4 - ◦ - - Fengjun Lv Consulting Fengjun Lv (Krizhevsky et al., 2012; Harel et al., 2007)”
Recognizing Image Style (2013, methodology): “We also model composition with a visual attention feature [11].”
Deep Level Sets for Salient Object Detection (2017, background): “Early works focused on fixation-level saliency detection [19, 14, 17] that aims to predict human’s attentional priority when viewing an image.”
Saliency detection using maximum symmetric surround (2010, methodology): “The saliency maps generated by several methods suffer from low resolution [7, 1, 14, 9, 11].”
Computational visual attention systems and their cognitive foundations: A survey (2010, methodology): “Other solutions are presented for example in [Itti et al. 1998; Itti and Koch 2001b; Harel et al. 2007].”
Quantitative Analysis of Human-Model Agreement in Visual Saliency Modeling: A Comparative Study (2013, neutral): “[26] 2006 I M W × H 18 HouCVPR: Hou et al.”
Modeling Search for People in 900 Scenes: A combined source model of eye guidance (2009, neutral): “Saccade onset was defined as the time when acceleration exceeded threshold and the saccade terminated when acceleration dropped below threshold.”
Hierarchical Image Saliency Detection on Extended CSSD (2016, neutral): “A few examples are shown in Fig.”
Pano2Vid: Automatic Cinematography for Watching 360° Videos (2016, methodology): “– Saliency trajectory—This baseline uses our pipeline, but replaces our discriminative capture-worthiness scores with the saliency score from a popular saliency method [30].”
A Stagewise Refinement Model for Detecting Salient Objects in Images (2017, background): “In [16, 15, 53, 23, 27, 38], saliency is measured by label propagation where initial labeling is propagated from the labeled elements to the unlabeled ones based on their pairwise affinities.”
New Contour Cue-Based Hybrid Sparse Learning for Salient Object Detection (2019, background): “Harel et al. [26] built a fully connected directed graph to mimic the neuron state transition process in the visual cortex.”
Salient Object Detection via Structured Matrix Decomposition (2017, background): “[40] who define a graph on image and adopt random walks to compute saliency.”
A unified approach to salient object detection via low rank matrix recovery (2012, methodology): “Compared with IT[12], MZ[18], IG[1], RC[4], GB[9], SR[10] and CA[8], our method achieves the best performance.”
Efficient, simultaneous detection of multi-class geospatial targets based on visual saliency modeling and discriminative learning of sparse coding (2014, methodology): “These models (Achanta et al., 2008, 2009; Borji, 2012; Han et al., 2006, 2011; Harel et al., 2006; Hou and Zhang, 2007; Itti et al., 1998; Judd et al., 2009; Zhang et al., 2008) generally assume that locations in the visual field that are distinctive from their contextual background are more likely ”
Context-aware saliency detection (2010, background): “The biologically-motivated algorithms for saliency estimation [9, 8, 24, 3, 6, 12] are based on principle (1).”
Non-local Deep Features for Salient Object Detection (2017, neutral): “The literature offers a wide variety of unsupervised methods working at the pixel level [2], the region level [11], with graph-based methods [15, 45], and with a Bayesian formulation [47].”
Shallow and Deep Convolutional Networks for Saliency Prediction (2016, background): “Notice how while GBVS [14] clearly outperforms our shallow convnet for MIT300 (Table 6), its results are much lower than our shallow convnet or Rare 2012 Improved [30] for the iSUN (Table 4) and SALICON datasets (Table 5).”
Saliency Detection on Light Field (2017, neutral): “These include algorithms based on spatiotemporal-cues (LC [41]), graph-based saliency (GB [42]), frequency-tuning (FT [39]), spectral residual (SS [43]), global-contrast (HC [26] and RC [6]), Low Rank Matrix Recovery (LRMR [20]), Graph-Based Manifold Ranking (GBMR [27]), focus-ness-based (UFO…”
Progressively Complementarity-Aware Fusion Network for RGB-D Salient Object Detection (2018, methodology): “Traditional saliency detection models [3, 7-10] are performed merely on RGB images and can be categorized as bottom-up and top-down pipelines.”