Saliency detection based on global and local short-term sparse representation

2016 ◽  
Vol 175 ◽  
pp. 81-89 ◽  
Author(s):  
Qiang Fan ◽  
Chun Qi
Keyword(s):  
Sparse Representation ◽  
Saliency Detection ◽  
Short Term ◽  
Global And Local

Saliency detection based on short-term sparse representation

2010 ◽  
Author(s):  
Xiaoshuai Sun ◽  
Hongxun Yao ◽  
Rongrong Ji ◽  
Pengfei Xu ◽  
Xianming Liu ◽  
...  
Keyword(s):  
Sparse Representation ◽  
Saliency Detection ◽  
Short Term

Sparse Representation of the Human Vision Information and the Saliency Detection Algorithm

2014 ◽  
Vol 513-517 ◽  
pp. 3349-3353
Author(s):  
Ju Bo Jin ◽  
Yu Xi Liu
Keyword(s):  
Firing Rate ◽  
Sparse Representation ◽  
Large Scale ◽  
Saliency Detection ◽  
Visual Saliency ◽  
Detection Algorithm ◽  
Human Vision ◽  
Short Term ◽  
Natural Statistics ◽  
Sparse Features

Representation and measurement are two important issues for saliency models. Different with previous works that learnt sparse features from large scale natural statistics, we propose to learn features from short-term statistics of single images. For saliency measurement, we defined basic firing rate (BFR) for each sparse feature, and then we propose to use feature activity rate (FAR) to measure the bottom-up visual saliency. The proposed FAR measure is biological plausible and easy to compute and with satisfied performance. Experiments on human trajectory positioning and psychological patterns demonstrate the effectiveness and robustness of our proposed method.

Saliency Detection: A Self-Adaption Sparse Representation Approach

2011 ◽  
Author(s):  
Gaoxiang Zhang ◽  
Feng Jiang ◽  
Debin Zhao ◽  
Xiaoshuai Sun ◽  
Shaohui Liu
Keyword(s):  
Sparse Representation ◽  
Saliency Detection

Bottom-up saliency detection with sparse representation of learnt texture atoms

2016 ◽  
Vol 60 ◽  
pp. 348-360 ◽  
Author(s):  
Mai Xu ◽  
Lai Jiang ◽  
Zhaoting Ye ◽  
Zulin Wang
Keyword(s):  
Sparse Representation ◽  
Saliency Detection ◽  
Bottom Up

PROGRESS AND REGRESS: UNDERSTANDING COMPLEX SOCIAL MEASURES AND THEIR TRADE-OFFS

2017 ◽  
Vol 34 (2) ◽  
pp. 164-189
Author(s):  
Daniel Austin Green ◽  
Roberta Q. Herzberg
Keyword(s):  
Income Inequality ◽  
Social Values ◽  
Scientific Progress ◽  
Well Being ◽  
Moral Progress ◽  
Short Term ◽  
Economic Progress ◽  
Trade Offs ◽  
Global And Local

Abstract:What is progress and what is not progress? We can talk about progress in lots of different arenas; we will focus primarily on economic and scientific progress, but also make brief reference to cultural and moral progress. In our discussion, we want to distinguish, especially, between overall, long-term progress and narrower, shorter-term progress or regress. We will refer to these as “global” and “local” progress, respectively. Of course, one can also regress; therefore, we will also look at instances where progress, along some dimension, slows or even moves backwards. Generally, such regress is local, and often still in a context of broader, global progress. In scientific progress, for example, there are many instances of short-term progress which, if not completely discarded or disproved, are at least substantially modified or fundamentally challenged. And yet, those research paths, even when later abandoned, still contributed to the overall progress of the field. In that sense, the regress (that is, rejection or modification of previous theories) is corrected by, but not in conflict with, the overall progress. In the case of economic progress, the concept of regress usually takes on a different form in which things that aren’t advancing progress don’t necessarily stop it, but are simply retarding progress — that is, making the rate of progress less efficient. The consequence, we suggest, is that when talking about economic progress, objections to certain consequences of economic progress (for instance, income inequality — a type of regress, in our terminology) should not be cordoned off and dealt with independently, but should be incorporated into the way we think about economic progress itself — as instances of local regress within a context of global progress. We explore the effects of these different relations between progress and regress to suggest some of the challenges those seeking to broaden the standard measure, GDP, to incorporate other social values of well-being will face moving forward.

scholarly journals Global and local contributions to surface curva- ture of healthy corneas

2012 ◽  
Vol 71 (4) ◽  
Author(s):  
Alan Rubin ◽  
Solani David Mathebula
Keyword(s):  
Physical Nature ◽  
Human Cornea ◽  
Short Term ◽  
Corneal Surface ◽  
Term Variation ◽  
Dioptric Power ◽  
Young Subjects ◽  
The Right ◽  
Global And Local ◽  
Healthy Eyes

This paper demonstrates for several healthy eyes the application of a simple model to understanding local and global contributions to short-term variation in anterior and posterior corneal curvature. Multiple axial anterior and posterior corneal radii and central corneal thicknesses for the right eyes of 10 young subjects were determined over time using a rotating Scheimpflug camera (Oculus Pentacam). The axial radii were transformed to corneal powers, and also to curvatures that were referred to a mid-corneal surface such that local and global contributions to short-term variation could be analyzed quantitatively.When variation of the anterior and posterior corneal surfaces of several healthy eyes are studied in terms of curvatures (rather than powers) it is the posterior surfaces that are more variable withthe global or macroscopic rather than local effects dominating. (Harris and Gillan found the same for an eye with mild keratoconus.) This finding is opposite to that when variation is considered in terms of dioptric power where the anterior corneal surface usually appears more variable. Possible reasons for this finding includes firstly that the posterior corneal surface has to be measured through the air-tear interface and anterior corneal surface,and thus some uncertainty in measurements of the posterior surface may relate to this limitation. Secondly, no attempt was made here to mathematically align the multiple surfaces as determined per eye and thus we cannot be certain that precisely the same central corneal region was measured each time.Investigators need to carefully consider whether they are more interested in the optical or physical nature of variation in surfaces such as the cornea since studies of the optical effects require theanalysis to be performed in terms of dioptric powers and  symmetric dioptric power space whereas studies of physical variation in the topography of the cornea and the possible reasons for such variability require the application of surface curvaturesin  surface curvature space. This paper describes the application and significance of both methods to facilitate understanding of short-term variation of the human cornea. It does not, however, attempt to make any definite claims as to what factors (seeabove) may be major contributors to such variability, and this complicated but interesting research issue requires further clarification. (S Afr Optom 2012 71(4) 146-158)

scholarly journals Combination of Spatial and Frequency Domains for Floating Object Detection on Complex Water Surfaces

2019 ◽  
Vol 9 (23) ◽  
pp. 5220
Author(s):  
Sun ◽  
Deng ◽  
Liu ◽  
Deng
Keyword(s):  
Object Detection ◽  
Saliency Detection ◽  
Small Sample ◽  
Low Rank ◽  
Support Vector ◽  
Frequency Domains ◽  
Floating Object ◽  
Texture Detection ◽  
Floating Objects ◽  
Global And Local

In order to address the problems of various interference factors and small sample acquisition in surface floating object detection, an object detection algorithm combining spatial and frequency domains is proposed. Firstly, a rough texture detection is performed in a spatial domain. A Fused Histogram of Oriented Gradient (FHOG) is combined with a Gray Level Co-occurrence Matrix (GLCM) to describe global and local information of floating objects, and sliding windows are classified by Support Vector Machines (SVM) with new texture features. Then, a novel frequency-based saliency detection method used in complex scenes is proposed. It adopts global and local low-rank decompositions to remove redundant regions caused by multiple interferences and retain floating objects. The final detection result is obtained by a strategy of combining bounding boxes from different processing domains. Experimental results show that the overall performance of the proposed method is superior to other popular methods, including traditional image segmentation, saliency detection, hand-crafted texture detection, and Convolutional Neural Network Based (CNN-based) object detection. The proposed method is characterized by small sample training and strong anti-interference ability in complex water scenes like ripple, reflection, and uneven illumination. The average precision of the proposed is 97.2%, with only 0.504 seconds of time consumption.

scholarly journals Investigating the network consequences of focal brain lesions through comparisons of real and simulated lesions

2021 ◽  
Vol 11 (1) ◽  
Author(s):  
Yuan Tao ◽  
Brenda Rapp
Keyword(s):  
Chronic Stroke ◽  
Simulation Studies ◽  
Short Term ◽  
Focal Lesions ◽  
Healthy Control ◽  
Quantitative Understanding ◽  
Network Properties ◽  
Computer Simulation Studies ◽  
Global And Local

AbstractGiven the increased interest in the functional human connectome, a number of computer simulation studies have sought to develop a better quantitative understanding of the effects of focal lesions on the brain’s functional network organization. However, there has been little work evaluating the predictions of this simulation work vis a vis real lesioned connectomes. One of the few relevant studies reported findings from real chronic focal lesions that only partially confirmed simulation predictions. We hypothesize that these discrepancies arose because although the effects of focal lesions likely consist of two components: short-term node subtraction and long-term network re-organization, previous simulation studies have primarily modeled only the short-term consequences of the subtraction of lesioned nodes and their connections. To evaluate this hypothesis, we compared network properties (modularity, participation coefficient, within-module degree) between real functional connectomes obtained from chronic stroke participants and “pseudo-lesioned” functional connectomes generated by subtracting the same sets of lesioned nodes/connections from healthy control connectomes. We found that, as we hypothesized, the network properties of real-lesioned connectomes in chronic stroke differed from those of the pseudo-lesioned connectomes which instantiated only the short-term consequences of node subtraction. Reflecting the long-term consequences of focal lesions, we found re-organization of the neurotopography of global and local hubs in the real but not the pseudo-lesioned connectomes. We conclude that the long-term network re-organization that occurs in response to focal lesions involves changes in functional connectivity within the remaining intact neural tissue that go well beyond the short-term consequences of node subtraction.

Sign in / Sign up

Export Citation Format

Share Document