Image Signature: Highlighting Sparse Salient Regions

By Xiaodi Hou, Jonathan Harel, Christof Koch. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2012

903 citations recorded by the source corpus.

Subfields: Cognitive Neuroscience, Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/48c2b8a35ce9468b84ac032af70cf2b100f5c1fc

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

Towards Explanation of DNN-based Prediction with Guided Feature Inversion (2018, methodology): “It shows that our method provides slightly better saliency detection result than SS [18], SIM [29] in terms of both two metrics.”
Visual Saliency Based on Scale-Space Analysis in the Frequency Domain (2013, neutral): “F”
Salient Object Detection via Structured Matrix Decomposition (2017, methodology): “427 Metric SMD ULR [26] SLR [28] LRR [27] GS [24] SF [50] CB [19] CA [7] SS [15] SEG [13] FT [49] SR [14] LC [16] WF↑ 0.”
Attentive Explanations: Justifying Decisions and Pointing to the Evidence (2016, background): “Saliency Map [20]: attends to the most salient parts of the image.”
Saliency detection by multi-context deep learning (2015, methodology): “In Table 2, we compare our approach with nine latest state-of-the-art methods, including IS [20], GBVS [17], SF [44], GC [12], CEOS [40], PCAS [41], GBMR [57], HS [56], and DRFI [25].”
Salient object detection: A survey (2014, methodology): “eration The attentive visual search mechanisms have been studied in different background and problem focuses: including salient object detection [8], [86], [115], [139], ﬁxation prediction [1], [18], [60], and object proposal generation [57], [64]– [67]. Salient object detection models usually aim t”
Two-Stage Learning to Predict Human Eye Fixations via SDAEs (2016, methodology): “with varying Gaussian blur standard deviation σ are usually convolved to smooth the obtained saliency maps as in [24] and [54] to show its influence on the shuffled-AUC score and help to find the optimal one for maximizing the performance of each saliency model.”
Salient Object Detection: A Benchmark (2015, background): “Since there exist many datasets that differ in number of images, number of objects per image, image resolution and annotation form (bounding box or accurate region mask), it is likely that models may rank differently across datasets.”
Creating Summaries from User Videos (2014, methodology): “We use the approach of [5] to predict a human attention score based on spatial [12] and temporal saliency (temporal gradients).”
Quantitative Analysis of Human-Model Agreement in Visual Saliency Modeling: A Comparative Study (2013, background): “For instance, it has been shown that smoothing the final saliency map of a model affects the scores [75].”
Saliency Detection: A Boolean Map Approach (2013, methodology): “On average, BMS, either using Lab or RGB, and the recent methods like AWS [10], ΔQDCT [33], CAS [11] and SigSal[14] give better performance than the others.”
KonIQ-10k: An Ecologically Valid Database for Deep Learning of Blind Image Quality Assessment (2019, methodology): “We combined the Viola-Jones face detector [48] over multiple poses (front, proﬁle left, and right) with a saliency detection method [47] into an importance map, see Fig.”
SALICON: Reducing the Semantic Gap in Saliency Prediction by Adapting Deep Neural Networks (2015, methodology): “We compare our method with 14 saliency prediction models , which are BMS [42], Adaptive Whitening Saliency (AWS) [11], Attention based on Information Maximization (AIM) [2], RARE [30], Local and Global saliency (LG) [1], eDN [39], Judd's model [21], image Signature Saliency (SigSal) [15], Context-Aw”
Learning to Predict Gaze in Egocentric Video (2013, methodology): “Other approaches also include spectral methods [9], sparse and efficient coding [10], graphical model [8] and learning based model [12].”
Large-Scale Optimization of Hierarchical Features for Saliency Prediction in Natural Images (2014, background): “Recently, spectrum-based methods demonstrated good performance despite low computational complexity [12, 11, 28].”