Competition for visual representation is an important mechanism for selective visual attention. The traditional global distinctiveness based saliency models usually compute the distinctiveness to measure saliency via comparing the difference of image patches in various spaces. In this paper, we propose to use an improved neural competition model to replace the comparison. The pairwise competition responses for a patch to all of the other patches are summed up to represent the distinctiveness of that patch. Particularly, the competition response is computed by a neural competition model with the dissimilarity bias and the gradient based feature inputs. Experimental results validate that the proposed model presents high effectiveness in saliency detection by outperforming nine state-of-the-art models.
The distinctiveness of image regions is widely used as the cue of saliency. Generally, the distinctiveness is computed according to the absolute difference of features. However, according to the image quality assessment (IQA) studies, the human visual system is highly sensitive to structural changes rather than absolute difference. Accordingly, we propose the computation of the structural dissimilarity between image patches as the distinctiveness measure for saliency detection. Similar to IQA models, the structural dissimilarity is computed based on the correlation of the structural features. The global structural dissimilarity of a patch to all the other patches represents saliency of the patch. We adopt two widely used structural features, namely the local contrast and gradient magnitude, into the structural dissimilarity computation in the proposed model. Without any postprocessing, the proposed model based on the correlation of either of the two structural features outperforms 11 state-of-the-art saliency models on three saliency databases.
The future video coding standard Versatile Video Coding (VVC) presents better encoding performance than the predecessor standards by employing a set of tools, including a multi-type tree block partition structure, more intra prediction directions, multiple transform functions, larger transform with high-frequency zeroing. Therefore, VVC could be more effective to remove the redundancy. Based on our experiments, the transform coefficients distribution (TCD) produced by the encoder of VVC would have a sharper peak. Particularly, the previously widely used Laplacian distribution and Cauchy distribution cannot fit the sharper TCD well. Note that the Laplacian distribution is included in the generalized Gaussian distribution (GGD) with the shape parameter equals one. Moreover, the smaller shape parameter will lead to a sharper peak. With this motivation, we propose to use the GGD with shape parameter equals 1/2, denoted as S/2 distribution, to model the sharper TCD of VVC. The experimental results show that the proposed S/2 distribution outperforms the widely used Laplacian and Cauchy distributions in terms of TCD fitting both in the main body and the tail parts. It also presents competitive performance with GGD, though there is only one parameter in S/2 distribution. We further propose a rate estimation model based on the S/2 distribution. The results show that the model based on the S/2 distribution is more accurate than one based on the Laplacian or Cauchy distribution in rate estimation.
We propose a novel local correlation based saliency model that is friendly to application of video coding. The
proposed model is developed in YCbCr color space. We extract feature maps with local mean and local contrast
of each channel image and its Gaussian blurred image, and produce rarity maps by calculating the correlation
between the feature maps of the original and blurred channels. The proposed saliency map is produced by a
combination of the local mean rarity maps and the local contrast rarity maps across all the channels. Experiments
validate that the proposed model works with excellent performance.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.