Saliency detection using quaternionic distance based weber local descriptor and level priors

Jian, Muwei; Qi, Qiang; Dong, Junyu; Sun, Xin; Sun, Yu; Lam, Kin‐Man

doi:10.1007/s11042-017-5032-z

Cited by 30 publications

(16 citation statements)

References 40 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…With the aim for localizing salient objects, a relevance feedback algorithm was employed to estimate weights for integrating textureinsensitive and region-based saliency measures [33]. Jian et al [34] proposed a saliency detection method by combining quaternionic distance based weber local descriptor and low-level priors. In [35], Li et al…”

Section: Related Workmentioning

confidence: 99%

Saliency detection based on directional patches extraction and principal local color contrast

Jian

Zhang

et al. 2018

Journal of Visual Communication and Image Representation

Self Cite

View full text Add to dashboard Cite

Saliency detection has become an active topic in both computer vision and multimedia fields. In this paper, we propose a novel computational model for saliency detection by integrating the holistic centerdirectional map with the principal local color contrast (PLCC) map. In the proposed framework, perceptual directional patches are firstly detected based on discrete wavelet frame transform (DWFT) and sparsity criterion, then the center of the spatial distribution of the extracted directional patches are utilized to locate the salient object in an image. Meanwhile, we proposed an efficient local color contrast method, called principal local color contrast (PLCC), to compute the color contrast between the salient object and the image background, which is sufficient to highlight and separate salient objects from complex background while dramatically reduce the computational cost. Finally, by incorporating the complementary visual cues of the global center-directional map with the PLCC map, a final compounded saliency map can be generated. Extensive experiments performed on three publicly available image databases, verify that the proposed scheme is able to achieve satisfactory results compared to other stateof-the-art saliency-detection algorithms.

show abstract

Section: Related Workmentioning

confidence: 99%

Saliency detection based on directional patches extraction and principal local color contrast

Jian

Zhang

et al. 2018

Journal of Visual Communication and Image Representation

Self Cite

View full text Add to dashboard Cite

show abstract

“…Apart from the dense detection framework using sliding windows scheme, like the HOG detector (Dalal and Triggs, 2005 ) and its modifications (Wang et al, 2009 ; Felzenszwalb et al, 2010 ; Yan et al, 2014 ; Pedersoli et al, 2015 ), there is another pipeline of detection methods using “attention” mechanism and is referred to as region-based detection methods (Girshick et al, 2014 ; Uijlings et al, 2013 ; Girshick, 2015 ; Jian et al, 2015 , 2017 ). These methods propose a number of high potential pedestrian candidate regions which is much less than that of sliding window methods.…”

Section: Related Workmentioning

confidence: 99%

Faster R-CNN for Robust Pedestrian Detection Using Semantic Segmentation Network

Liu

Stathaki

2018

Front. Neurorobot.

View full text Add to dashboard Cite

Convolutional neural networks (CNN) have enabled significant improvements in pedestrian detection owing to the strong representation ability of the CNN features. However, it is generally difficult to reduce false positives on hard negative samples such as tree leaves, traffic lights, poles, etc. Some of these hard negatives can be removed by making use of high level semantic vision cues. In this paper, we propose a region-based CNN method which makes use of semantic cues for better pedestrian detection. Our method extends the Faster R-CNN detection framework by adding a branch of network for semantic image segmentation. The semantic network aims to compute complementary higher level semantic features to be integrated with the convolutional features. We make use of multi-resolution feature maps extracted from different network layers in order to ensure good detection accuracy for pedestrians at different scales. Boosted forest is used for training the integrated features in a cascaded manner for hard negatives mining. Experiments on the Caltech pedestrian dataset show improvements on detection accuracy with the semantic network. With the deep VGG16 model, our pedestrian detection method achieves robust detection performance on the Caltech dataset.

show abstract

“…Unlike local methods, which are sensitive to high frequency image contents like edges and noise, global methods are less effective when the textured regions of salient objects are similar to the background. In order to reduce the effect of background and generate saliency maps with little noise, QDWD, which was initially proposed to detect the outliers and edges in an image [31], is utilized to represent the global shape information for the HVS to detect saliency [19,35]. A quaternion q is made up of one real part and three imaginary parts, as follows: D q q using the method in [31].…”

Section: Quaternionic Distance Based Weber Descriptormentioning

confidence: 99%

Saliency Detection via Combining Global Shape and Local Cue Estimation

Jian

Yin

et al. 2017

Lecture Notes in Computer Science

Self Cite

View full text Add to dashboard Cite

Recently, saliency detection has become a hot issue in computer vision. In this paper, a novel framework for image saliency detection is introduced by modeling global shape and local cue estimation simultaneously. Firstly, Quaternionic Distance Based Weber Descriptor (QDWD), which was initially designed for detecting outliers in color images, is used to model the salient object shape in an image. Secondly, we detect local saliency based on the reconstruction error by using a locality-constrained linear coding algorithm. Finally, by integrating global shape with local cue, a reliable saliency map can be computed and estimated. Experimental results, based on two widely used and openly available databases, show that the proposed method can produce reliable and promising results, compared to other state-of-the-art saliencydetection algorithms.

show abstract

Saliency detection using quaternionic distance based weber local descriptor and level priors

Cited by 30 publications

References 40 publications

Saliency detection based on directional patches extraction and principal local color contrast

Saliency detection based on directional patches extraction and principal local color contrast

Faster R-CNN for Robust Pedestrian Detection Using Semantic Segmentation Network

Saliency Detection via Combining Global Shape and Local Cue Estimation

Contact Info

Product

Resources

About