Home Physical Sciences The visual attention saliency map for movie retrospection
Article Open Access

The visual attention saliency map for movie retrospection

  • Anna Rogalska EMAIL logo and Piotr Napieralski
Published/Copyright: April 18, 2018
Download Article (PDF)
Open Physics
From the journal Open Physics Volume 16 Issue 1

Abstract

The visual saliency map is becoming important and challenging for many scientific disciplines (robotic systems, psychophysics, cognitive neuroscience and computer science). Map created by the model indicates possible salient regions by taking into consideration face presence and motion which is essential in motion pictures. By combining we can obtain credible saliency map with a low computational cost.

Keywords: saliency prediction; visual attention; gaze tracking; computer graphics
PACS: 42.30.-d; 87.19.lt

1 Introduction

The film and motion picture industry is revenue of the forecast to increase nearly to 50 billion U.S. dollar in 2020 [1]. While the US system is profitable, other film industries tend to be financed through government subsidies. It is important not to exceed the movie budget, while maintaining desired picture quality. That is why it is important to review pre-production scenes. The proposed solution allows for the verification if a certain image region visually stands out, compared to its surrounding area, important for the audience (visual saliency). Saliency-extraction techniques of this region generate a saliency map initially presented by Itti et al. [2]. Saliency map indicates regions that are likely to attract human attention. Red regions indicate high saliency level, blue regions indicate low saliency level. Figure 1 shows an example saliency map (presented as a heatmap for legibility).

Figure 1 Left: original frame, right: saliency map of frame
Figure 1

Left: original frame, right: saliency map of frame

Nowadays domain of visual attenuation has evolved in the many scientific disciplines (robotic systems, psychophysics, cognitive neuroscience and computer science).

Visual saliency detection is widely used in video compression, image retargeting, video retargeting or object detection [3]. Saliency-detection methods, combine a probability of saliency with spatial variance of color of a superpixel, global contrast map or spatially-aware (non-global) color quantization [4]. The final saliency map is calculated by linearly interpolating the product of these components. Choice of these components is made by the fact that some cues take more attention in the observed scene. Human Visual System (HVS) is sensitive to features like changes in color, specific shapes, light intensity or human faces [5]. These features indicate Regions Of Interest (ROI). Such regions, with certain probability, are likely to attract attention. According to Yarbus [6] the task that viewers have will also influence ROI. For instance, when a viewer was asked to simply look at the picture (figure 2) his gaze track looked different compared to the gaze track after questions like “How old are people in the room?” or “How are they dressed?”.

Figure 2 Different gaze path for the same picture and different tasks [6]
Figure 2

Different gaze path for the same picture and different tasks [6]

These differences in gaze track are due to HVS being able to pay attention to only limited number of ROI at certain point of time.

When attention cues are accumulated in the picture, some of them are more likely to attract attention than the others. For example when there is a girl dressed in yellow raincoat what attracts attention first is her face and later – the yellow raincoat (contrast, intensity). This becomes more complicated where motion picture is analyzed instead of a static one. Duration of shot influences ROI, because some attention cues attract gaze earlier than the other. Referring to previous example, when the shot is only two seconds long, only girl’s face will attract attention. If the shot is longer, then after a while viewers will look at the rest of the scene (for example at yellow raincoat).

Popularity of stereoscopic movies, bring metrics of visual comfort for stereoscopic content [7, 8, 9]. These methods are far more complicated as they take into account disparity (which can itself lead to loss in visual comfort [10]), planar and depth velocities, as well as the spatial frequency of luminance contrast and motion on the screen plane. The saliency stereoscopy models take into account more components than an ordinary picture. This model is made by fusing depth perception and visual comfort factors, like image and visual comfort based saliency, and above that depth saliency.

2 Existing methods

Saliency prediction models can be divided into two basic categories: bottom-up and top-down. Bottom-up means that we analyze an image but not what is in the image. For example color, intensity and orientation are bottom-up cues, while face presence, movement, interesting or emotional objects are top-down cues. There already exist several bottom-up models such as Itti et al. [2] which is based on color, intensity and orientation and combines multiscale image features into a single topographical saliency map, Levine et al. [11] combines local contrast with global contrast in order to detect both small and large saliency regions, detect saliency in cluttered scenes and inhibit repeating patterns. Yildrim et al. [4] combines global contrast with probability of saliency which is based on spatial center and variances of colors. These models include only bottom-up features and ignore top-down ones. This can lead to building inaccurate saliency maps. Several top-down models, such as: Xu et al. [12] proposed not only face-based saliency map, but also facial features-based saliency map. He verified that the fixations tend to cluster around facial features, when viewing images with large faces. Yu et al. [15] predicted saliency maps by indicating interesting objects using a novel object extraction approach by integrating sketch-like and envelope-like maps.

Unfortunately bottom-up features cannot be ignored as they attract attention during first 500 milliseconds. Because of that, there were developed models that combine both bottom-up and top-down features. For example Xu et at. [16] combined Ittis model with movement by using adaptive fusion method that merges the spatial and temporal saliency maps of each frame into its spatiotemporal saliency map. Kucerowa et al. [5] chose texture and faces as attention cues and combined them using local context suppression of multiple cues.

Most existing models take into consideration static images. However motion picture plays an essential role in film industry. It may turn out that some attention cues are far more important when it comes to motion picture. For example it is arguable if face will attract more attention than contrast or movement.

Another thing is that some of presented models are not versatile. As an example: method presented by Kucerowa et al. [5] will not work for images without text or face, as saliency map is based only on face and text presence.

3 Method description

We propose a model that is versatile and can be used in motion pictures. We work on videos with natural scenes, with or without faces. Proposed model, compared to state-of-the-art methods was improved by adding more attention cues. Based on the eye tracking measurements [13], proposed model evaluates the level of importance regions associated to viewing movie or motion picture [14]. We create saliency map by combining four attention cues:

  1. Global contrast

  2. Distance from the center of an image

  3. Face

  4. Movement

We combine these features into a single saliency map using following equation:

Si=w1∗Ki+w2∗Gi+w3∗Fi+w4∗Miw1+w2+w3+w4=1w1,w2,w3,w4−mapweightsi−imagepixelK−contrastG−distancefromthecenterofanimageF−facepresenceM−movementS−finalsaliencymap

Using this equation we can indicate which attention cues are more important than others, so the final saliency map can look differently depending on weight parameters. In the following section we elaborate on the importance of weight parameters. We also test how different weight parameters influence saliency map accuracy.

3.1 Global contrast

We use global contrast instead of local contrast for improved efficiency. We construct histogram and calculate difference for each color.

Ki=∑j=1mfjCi−CjKi−saliencyvalueforcolorm−numberofcolorsC−colorvaluefj−frequencyofcolorCj

3.2 Distance from the center of an image

Research has showed that the probability of salient object appearance decreases with the distance from the center of an image. We use Gaussian function to model this.

Gx,y=12πσ2e−x2+y22σ2x,y−coordinatesofpixelwithrespecttoimagecenter

3.3 Face

In this case, research has shown that face is first to attract attention during first 500 milliseconds of viewing (as observed in 80% of cases). For face detection we use Haar clasificatior.

3.4 Movement

In order to detect motion we use optical flow, using Gunnar Farnebacks [19] algorithm. We assume that pixel intensity does not change in consecutive frame. We calculate the distance between a pixel in the first and in the second frame.

Mx,y,t=Mx+Δx,y+Δy,t+Δti−imagepixelx,y−pixelcoordinates

4 Experimental results

We compared our saliency map with ground truth map, which is comprised of human eye fixation map. The above methods were tested on 1920 × 1080px, 1m 42s long video which is Isyrius company marketing material. The video consists of different types of scenes with variety of color schemes, different dynamics and shot length. This particular video was chosen because of its dynamic montage, as well as, shot richness and variety. We obtained eyetracking data by using Tobii eyetracker and tested 18 persons aged from 22 to 37. On computer with Intel Core i7 processor and 8GB RAM the saliency map was calculated after 210 minutes. That gives 0.2 fps processing speed. Figure 3 shows a schematic view of our setup.

Figure 3 Schematic view of research stand
Figure 3

Schematic view of research stand

We tested accuracy of proposed method by comparing calculated saliency map SA (its black and white version) with ground truth map SB, using Linear Correlation Coefficient (LCC):

LCC=∑i=1nSAi−SA¯SBi−SB¯∑i=1nSAi−SA¯2∑i=1nSBi−SB¯2n−numberofimagepixelsi−pixelSA−calculatedsaliencymapSB−groundtruthmap

Figure 4 shows scalculated saliency map SA and ground truth map (eyetracking map) SB for an example frame.

Figure 4 Left: calculated saliency map SA, right: ground truth map SB
Figure 4

Left: calculated saliency map SA, right: ground truth map SB

LCC was calculated for each frame and averaged for the whole video clip. Figure 5 shows LCC for our method compared to the other existing methods.

Figure 5 Comparison results for different saliency models
Figure 5

Comparison results for different saliency models

Figure 6 shows four different saliency maps (right column). The first map from the top is a global contrast based saliency map. Another map is generated by distance from center attention cue. Third map represents face presence based saliency map. The last map is motion based saliency map. In the bottom-left corner eyetracking map (overlayed on original frame) is shown. In the top-left corner saliency map is presented as a heatmap. This heatmap was generated using weights: contrast 30%, distance from center 20%, face 20%, motion 30%. Following the above, there is another saliency map (generated with different weights: contrast 10%, distance from center 20%, face 20%, motion 50%), and further down to the right is the original footage. Such presentation shows how each of attention cues influences final saliency map. Moreover, it can be compared at the same time with actual fixation points.

Figure 6 Right column (from top):. global contrast based saliency map, distance from center based saliency map, face presence based saliency map, motion based saliency map
Figure 6

Right column (from top):. global contrast based saliency map, distance from center based saliency map, face presence based saliency map, motion based saliency map

As can be seen in figure 6, different attention cues generate completely different saliency maps. Looking at figure 7 we can also see that they are also responsible for different LCC values. Figure 7 depicts how each attention cue influence LCC (each saliency map was calculated for every frame of the video). For video used for this research, LCC with use of three maps has value of 0.68. As we can see in figure 7 each of maps alone provides less LCC value. That indicates that combing them in a right way (with optimal weights) will generate credible saliency map.

Figure 7 Comparison results for different attention cues
Figure 7

Comparison results for different attention cues

5 Conclusion

Our method can be very useful in the film industry. It indicates possible salient regions by taking into consideration global contrast, distance from the center of an image, face presence and motion. Our method is versatile and suitable for fullHD videos. In contrast, previous methods considered mostly robust static features and were computationally complex so they are not suitable for videos. Also, many state-of-the-art methods can be completely misleading in motion picture as they neglect the importance of different attention cues, which we take into consideration.

We carried out several tests that gave us eyetracking data which we could compare with generated saliency maps. These data showed the deficiency in our method and at the same time provided us with interesting observations concerning relations between different attention cues.

Future work considers designating optimal weight parameters when combining obtained four saliency maps into final saliency map, adding more attention cues (specific for film industry) as horizon line and golden proportion, and improving processing speed.

References

[1] The Statistics Portal, https://www.statista.com, 20.04.2017Search in Google Scholar

[2] Itti L., Koch C., Niebur E., A model of saliency-based visual attention for rapid scene analysis., IEEE Transactions on PAMI 20, 199810.1109/34.730558Search in Google Scholar

[3] Puchala D., Yatsymirskyy M., Joint Compression And Encryption of Visual Data Using Orthogonal Parametric Transforms, Bulletin of the Polish Academy of Sciences-Technical Sciences, 2016, 64, 2, 373-382.10.1515/bpasts-2016-0042Search in Google Scholar

[4] Yildirim G., Susstrunk S., FASA: Fast, Accurate, and Size-Aware Salient Object Detection, Computer Vision, 201510.1007/978-3-319-16811-1_34Search in Google Scholar

[5] Kucerova J., Sikudova E., Saliency map augmentation with facial detection, Proceedings of CESCG, 2011Search in Google Scholar

[6] Yarbus A.L., Eye Movements and Vision, Plenum, New York, 1967.10.1007/978-1-4899-5379-7Search in Google Scholar

[7] Du S., Masia B., Hu S., Gutierrez D., A metric of visual comfort for stereoscopic motion, ACM Trans. Graph., 2013, 32, 6, 22210.1145/2508363.2508387Search in Google Scholar

[8] Pollock B.T., Burton M., Kelly J.W., Gilbert S., Winer E., The Right View from the Wrong Location: Depth Perception in Stereoscopic Multi-User Virtual Environments, IEEE Transactions on Visualization and Computer Graphics, 2012, 18, 581-588.10.1109/TVCG.2012.58Search in Google Scholar PubMed

[9] Qiuping J., Feng Sh., Gangyi J., Mei Y., Zongju P., Changhong Y., A depth perception and visual comfort guided computational model for stereoscopic 3D visual saliency, Signal Processing Image Communication, 2015Search in Google Scholar

[10] Kowalczyk M., Napieralski P., Noise resistant method enabling detection of vertical disparity in three-dimensional visualizations, 2017 18th International Symposium on Electromagnetic Fields in Mechatronics, Electrical and Electronic Engineering (ISEF) Book of Abstracts, Lodz, Poland, 2017, 1-2, 10.1109/ISEF.2017.8090694Search in Google Scholar

[11] Levine M., An X., He H., Saliency detection based on frequency and spatial domain analyses, Proceedings of the BMVC, 2011Search in Google Scholar

[12] Xu M., Ren Y., Wang Z., Learning to Predict Saliency on Face Images, IEEE International Conference on Computer Vision (ICCV), 2015, 3907-391510.1109/ICCV.2015.445Search in Google Scholar

[13] Wojciechowski A., Fornalczyk K., Single web camera robust interactive eye-gaze tracking method, Bulletin of the Polish Academy of Sciences-Technical Sciences 2015, 63, 879-886.10.1515/bpasts-2015-0100Search in Google Scholar

[14] Rogalska A., Napieralski P., A model of saliency-based visual attention for movie retrospection, 2017 18th International Symposium on Electromagnetic Fields in Mechatronics, Electrical and Electronic Engineering (ISEF) Book of Abstracts, Lodz, Poland, 2017, 1-2, 10.1109/ISEF.2017.8090692Search in Google Scholar

[15] Yu H., Li J., Tian Y., Huang T., Automatic interesting object extraction from images using complementary saliency maps, Proceedings of the 18th ACM international conference on Multimedia, 201010.1145/1873951.1874105Search in Google Scholar

[16] Xu J., Tu Q., Li C., Gao R., Men A., Video saliency map detection based on global motion estimation, IEEE International Conference on Multimedia and Expo Workshops (ICMEW), 2015, 1-6.Search in Google Scholar

[17] Farneback G., Two-frame motion estimation based on polynomial expansion, Lecture Notes in Computer Science, 2003, 363-370.10.1007/3-540-45103-X_50Search in Google Scholar

[18] Cheng M.M., Mitra N.J., Huang X., Torr P.H.S., Hu S.M., Global Contrast Based Salient Region Detection, IEEE Transactions on Pattern Analysis and Machine Intelligence, 2015, 37, 3, 569-582.10.1109/TPAMI.2014.2345401Search in Google Scholar PubMed

[19] Borji A., Boosting bottom-up and top-down visual features for saliency estimation, IEEE Conference on Computer Vision and Pattern Recognition, 201210.1109/CVPR.2012.6247706Search in Google Scholar

[20] Ko B., Nam J., Object-of-interest image segmentation based on human attention and semantic region clustering, J. Opt. Soc. Am., 2006, 23(10), 246210.1364/JOSAA.23.002462Search in Google Scholar

[21] Liu T., Yuan Z., Sun J., Wang J., Zheng N., Tang X., Shum H., Learning to detect a salient object, CVPR, 2007, 1-8.10.1109/CVPR.2007.383047Search in Google Scholar

[22] Judd T., Ehinger K., Durand F., Torralba A., Learning to predict where humans look, ICCV, 200910.1109/ICCV.2009.5459462Search in Google Scholar

[23] Subramanian R., Katti H., Sebe N., Kankanhalli M., Chua T.S., An eye fixation database for saliency detection in images, ECCV, 2010Search in Google Scholar

Received: 2017-11-04
Accepted: 2017-11-24
Published Online: 2018-04-18

© 2018 A. Rogalska and P. Napieralski, published by De Gruyter

This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 License.

Articles in the same Issue

  1. Regular Articles
  2. A modified Fermi-Walker derivative for inextensible flows of binormal spherical image
  3. Algebraic aspects of evolution partial differential equation arising in the study of constant elasticity of variance model from financial mathematics
  4. Three-dimensional atom localization via probe absorption in a cascade four-level atomic system
  5. Determination of the energy transitions and half-lives of Rubidium nuclei
  6. Three phase heat and mass transfer model for unsaturated soil freezing process: Part 1 - model development
  7. Three phase heat and mass transfer model for unsaturated soil freezing process: Part 2 - model validation
  8. Mathematical model for thermal and entropy analysis of thermal solar collectors by using Maxwell nanofluids with slip conditions, thermal radiation and variable thermal conductivity
  9. Constructing analytic solutions on the Tricomi equation
  10. Feynman diagrams and rooted maps
  11. New type of chaos synchronization in discrete-time systems: the F-M synchronization
  12. Unsteady flow of fractional Oldroyd-B fluids through rotating annulus
  13. A note on the uniqueness of 2D elastostatic problems formulated by different types of potential functions
  14. On the conservation laws and solutions of a (2+1) dimensional KdV-mKdV equation of mathematical physics
  15. Computational methods and traveling wave solutions for the fourth-order nonlinear Ablowitz-Kaup-Newell-Segur water wave dynamical equation via two methods and its applications
  16. Siewert solutions of transcendental equations, generalized Lambert functions and physical applications
  17. Numerical solution of mixed convection flow of an MHD Jeffery fluid over an exponentially stretching sheet in the presence of thermal radiation and chemical reaction
  18. A new three-dimensional chaotic flow with one stable equilibrium: dynamical properties and complexity analysis
  19. Dynamics of a dry-rebounding drop: observations, simulations, and modeling
  20. Modeling the initial mechanical response and yielding behavior of gelled crude oil
  21. Lie symmetry analysis and conservation laws for the time fractional simplified modified Kawahara equation
  22. Solitary wave solutions of two KdV-type equations
  23. Applying industrial tomography to control and optimization flow systems
  24. Reconstructing time series into a complex network to assess the evolution dynamics of the correlations among energy prices
  25. An optimal solution for software testing case generation based on particle swarm optimization
  26. Optimal system, nonlinear self-adjointness and conservation laws for generalized shallow water wave equation
  27. Alternative methods for solving nonlinear two-point boundary value problems
  28. Global model simulation of OH production in pulsed-DC atmospheric pressure helium-air plasma jets
  29. Experimental investigation on optical vortex tweezers for microbubble trapping
  30. Joint measurements of optical parameters by irradiance scintillation and angle-of-arrival fluctuations
  31. M-polynomials and topological indices of hex-derived networks
  32. Generalized convergence analysis of the fractional order systems
  33. Porous flow characteristics of solution-gas drive in tight oil reservoirs
  34. Complementary wave solutions for the long-short wave resonance model via the extended trial equation method and the generalized Kudryashov method
  35. A Note on Koide’s Doubly Special Parametrization of Quark Masses
  36. On right-angled spherical Artin monoid of type Dn
  37. Gas flow regimes judgement in nanoporous media by digital core analysis
  38. 4 + n-dimensional water and waves on four and eleven-dimensional manifolds
  39. Stabilization and Analytic Approximate Solutions of an Optimal Control Problem
  40. On the equations of electrodynamics in a flat or curved spacetime and a possible interaction energy
  41. New prediction method for transient productivity of fractured five-spot patterns in low permeability reservoirs at high water cut stages
  42. The collinear equilibrium points in the restricted three body problem with triaxial primaries
  43. Detection of the damage threshold of fused silica components and morphologies of repaired damage sites based on the beam deflection method
  44. On the bivariate spectral quasi-linearization method for solving the two-dimensional Bratu problem
  45. Ion acoustic quasi-soliton in an electron-positron-ion plasma with superthermal electrons and positrons
  46. Analysis of projectile motion in view of conformable derivative
  47. Computing multiple ABC index and multiple GA index of some grid graphs
  48. Terahertz pulse imaging: A novel denoising method by combing the ant colony algorithm with the compressive sensing
  49. Characteristics of microscopic pore-throat structure of tight oil reservoirs in Sichuan Basin measured by rate-controlled mercury injection
  50. An activity window model for social interaction structure on Twitter
  51. Transient thermal regime trough the constitutive matrix applied to asynchronous electrical machine using the cell method
  52. On the zagreb polynomials of benzenoid systems
  53. Integrability analysis of the partial differential equation describing the classical bond-pricing model of mathematical finance
  54. The Greek parameters of a continuous arithmetic Asian option pricing model via Laplace Adomian decomposition method
  55. Quantifying the global solar radiation received in Pietermaritzburg, KwaZulu-Natal to motivate the consumption of solar technologies
  56. Sturm-Liouville difference equations having Bessel and hydrogen atom potential type
  57. Study on the response characteristics of oil wells after deep profile control in low permeability fractured reservoirs
  58. Depiction and analysis of a modified theta shaped double negative metamaterial for satellite application
  59. An attempt to geometrize electromagnetism
  60. Structure of traveling wave solutions for some nonlinear models via modified mathematical method
  61. Thermo-convective instability in a rotating ferromagnetic fluid layer with temperature modulation
  62. Construction of new solitary wave solutions of generalized Zakharov-Kuznetsov-Benjamin-Bona-Mahony and simplified modified form of Camassa-Holm equations
  63. Effect of magnetic field and heat source on Upper-convected-maxwell fluid in a porous channel
  64. Physical cues of biomaterials guide stem cell fate of differentiation: The effect of elasticity of cell culture biomaterials
  65. Shooting method analysis in wire coating withdrawing from a bath of Oldroyd 8-constant fluid with temperature dependent viscosity
  66. Rank correlation between centrality metrics in complex networks: an empirical study
  67. Special Issue: The 18th International Symposium on Electromagnetic Fields in Mechatronics, Electrical and Electronic Engineering
  68. Modeling of electric and heat processes in spot resistance welding of cross-wire steel bars
  69. Dynamic characteristics of triaxial active control magnetic bearing with asymmetric structure
  70. Design optimization of an axial-field eddy-current magnetic coupling based on magneto-thermal analytical model
  71. Thermal constitutive matrix applied to asynchronous electrical machine using the cell method
  72. Temperature distribution around thin electroconductive layers created on composite textile substrates
  73. Model of the multipolar engine with decreased cogging torque by asymmetrical distribution of the magnets
  74. Analysis of spatial thermal field in a magnetic bearing
  75. Use of the mathematical model of the ignition system to analyze the spark discharge, including the destruction of spark plug electrodes
  76. Assessment of short/long term electric field strength measurements for a pilot district
  77. Simulation study and experimental results for detection and classification of the transient capacitor inrush current using discrete wavelet transform and artificial intelligence
  78. Magnetic transmission gear finite element simulation with iron pole hysteresis
  79. Pulsed excitation terahertz tomography – multiparametric approach
  80. Low and high frequency model of three phase transformer by frequency response analysis measurement
  81. Multivariable polynomial fitting of controlled single-phase nonlinear load of input current total harmonic distortion
  82. Optimal design of a for middle-low-speed maglev trains
  83. Eddy current modeling in linear and nonlinear multifilamentary composite materials
  84. The visual attention saliency map for movie retrospection
  85. AC/DC current ratio in a current superimposition variable flux reluctance machine
  86. Influence of material uncertainties on the RLC parameters of wound inductors modeled using the finite element method
  87. Cogging force reduction in linear tubular flux switching permanent-magnet machines
  88. Modeling hysteresis curves of La(FeCoSi)13 compound near the transition point with the GRUCAD model
  89. Electro-magneto-hydrodynamic lubrication
  90. 3-D Electromagnetic field analysis of wireless power transfer system using K computer
  91. Simplified simulation technique of rotating, induction heated, calender rolls for study of temperature field control
  92. Design, fabrication and testing of electroadhesive interdigital electrodes
  93. A method to reduce partial discharges in motor windings fed by PWM inverter
  94. Reluctance network lumped mechanical & thermal models for the modeling and predesign of concentrated flux synchronous machine
  95. Special Issue Applications of Nonlinear Dynamics
  96. Study on dynamic characteristics of silo-stock-foundation interaction system under seismic load
  97. Microblog topic evolution computing based on LDA algorithm
  98. Modeling the creep damage effect on the creep crack growth behavior of rotor steel
  99. Neighborhood condition for all fractional (g, f, n′, m)-critical deleted graphs
  100. Chinese open information extraction based on DBMCSS in the field of national information resources
  101. 10.1515/phys-2018-0079
  102. CPW-fed circularly-polarized antenna array with high front-to-back ratio and low-profile
  103. Intelligent Monitoring Network Construction based on the utilization of the Internet of things (IoT) in the Metallurgical Coking Process
  104. Temperature detection technology of power equipment based on Fiber Bragg Grating
  105. Research on a rotational speed control strategy of the mandrel in a rotary steering system
  106. Dynamic load balancing algorithm for large data flow in distributed complex networks
  107. Super-structured photonic crystal fiber Bragg grating biosensor image model based on sparse matrix
  108. Fractal-based techniques for physiological time series: An updated approach
  109. Analysis of the Imaging Characteristics of the KB and KBA X-ray Microscopes at Non-coaxial Grazing Incidence
  110. Application of modified culture Kalman filter in bearing fault diagnosis
  111. Exact solutions and conservation laws for the modified equal width-Burgers equation
  112. On topological properties of block shift and hierarchical hypercube networks
  113. Elastic properties and plane acoustic velocity of cubic Sr2CaMoO6 and Sr2CaWO6 from first-principles calculations
  114. A note on the transmission feasibility problem in networks
  115. Ontology learning algorithm using weak functions
  116. Diagnosis of the power frequency vacuum arc shape based on 2D-PIV
  117. Parametric simulation analysis and reliability of escalator truss
  118. A new algorithm for real economy benefit evaluation based on big data analysis
  119. Synergy analysis of agricultural economic cycle fluctuation based on ant colony algorithm
  120. Multi-level encryption algorithm for user-related information across social networks
  121. Multi-target tracking algorithm in intelligent transportation based on wireless sensor network
  122. Fast recognition method of moving video images based on BP neural networks
  123. Compressed sensing image restoration algorithm based on improved SURF operator
  124. Design of load optimal control algorithm for smart grid based on demand response in different scenarios
  125. Face recognition method based on GA-BP neural network algorithm
  126. Optimal path selection algorithm for mobile beacons in sensor network under non-dense distribution
  127. Localization and recognition algorithm for fuzzy anomaly data in big data networks
  128. Urban road traffic flow control under incidental congestion as a function of accident duration
  129. Optimization design of reconfiguration algorithm for high voltage power distribution network based on ant colony algorithm
  130. Feasibility simulation of aseismic structure design for long-span bridges
  131. Construction of renewable energy supply chain model based on LCA
  132. The tribological properties study of carbon fabric/ epoxy composites reinforced by nano-TiO2 and MWNTs
  133. A text-Image feature mapping algorithm based on transfer learning
  134. Fast recognition algorithm for static traffic sign information
  135. Topical Issue: Clean Energy: Materials, Processes and Energy Generation
  136. An investigation of the melting process of RT-35 filled circular thermal energy storage system
  137. Numerical analysis on the dynamic response of a plate-and-frame membrane humidifier for PEMFC vehicles under various operating conditions
  138. Energy converting layers for thin-film flexible photovoltaic structures
  139. Effect of convection heat transfer on thermal energy storage unit
https://doi.org/10.1515/phys-2018-0027
Keywords for this article
saliency prediction; visual attention; gaze tracking; computer graphics
Creative Commons
BY-NC-ND 4.0

Articles in the same Issue

  1. Regular Articles
  2. A modified Fermi-Walker derivative for inextensible flows of binormal spherical image
  3. Algebraic aspects of evolution partial differential equation arising in the study of constant elasticity of variance model from financial mathematics
  4. Three-dimensional atom localization via probe absorption in a cascade four-level atomic system
  5. Determination of the energy transitions and half-lives of Rubidium nuclei
  6. Three phase heat and mass transfer model for unsaturated soil freezing process: Part 1 - model development
  7. Three phase heat and mass transfer model for unsaturated soil freezing process: Part 2 - model validation
  8. Mathematical model for thermal and entropy analysis of thermal solar collectors by using Maxwell nanofluids with slip conditions, thermal radiation and variable thermal conductivity
  9. Constructing analytic solutions on the Tricomi equation
  10. Feynman diagrams and rooted maps
  11. New type of chaos synchronization in discrete-time systems: the F-M synchronization
  12. Unsteady flow of fractional Oldroyd-B fluids through rotating annulus
  13. A note on the uniqueness of 2D elastostatic problems formulated by different types of potential functions
  14. On the conservation laws and solutions of a (2+1) dimensional KdV-mKdV equation of mathematical physics
  15. Computational methods and traveling wave solutions for the fourth-order nonlinear Ablowitz-Kaup-Newell-Segur water wave dynamical equation via two methods and its applications
  16. Siewert solutions of transcendental equations, generalized Lambert functions and physical applications
  17. Numerical solution of mixed convection flow of an MHD Jeffery fluid over an exponentially stretching sheet in the presence of thermal radiation and chemical reaction
  18. A new three-dimensional chaotic flow with one stable equilibrium: dynamical properties and complexity analysis
  19. Dynamics of a dry-rebounding drop: observations, simulations, and modeling
  20. Modeling the initial mechanical response and yielding behavior of gelled crude oil
  21. Lie symmetry analysis and conservation laws for the time fractional simplified modified Kawahara equation
  22. Solitary wave solutions of two KdV-type equations
  23. Applying industrial tomography to control and optimization flow systems
  24. Reconstructing time series into a complex network to assess the evolution dynamics of the correlations among energy prices
  25. An optimal solution for software testing case generation based on particle swarm optimization
  26. Optimal system, nonlinear self-adjointness and conservation laws for generalized shallow water wave equation
  27. Alternative methods for solving nonlinear two-point boundary value problems
  28. Global model simulation of OH production in pulsed-DC atmospheric pressure helium-air plasma jets
  29. Experimental investigation on optical vortex tweezers for microbubble trapping
  30. Joint measurements of optical parameters by irradiance scintillation and angle-of-arrival fluctuations
  31. M-polynomials and topological indices of hex-derived networks
  32. Generalized convergence analysis of the fractional order systems
  33. Porous flow characteristics of solution-gas drive in tight oil reservoirs
  34. Complementary wave solutions for the long-short wave resonance model via the extended trial equation method and the generalized Kudryashov method
  35. A Note on Koide’s Doubly Special Parametrization of Quark Masses
  36. On right-angled spherical Artin monoid of type Dn
  37. Gas flow regimes judgement in nanoporous media by digital core analysis
  38. 4 + n-dimensional water and waves on four and eleven-dimensional manifolds
  39. Stabilization and Analytic Approximate Solutions of an Optimal Control Problem
  40. On the equations of electrodynamics in a flat or curved spacetime and a possible interaction energy
  41. New prediction method for transient productivity of fractured five-spot patterns in low permeability reservoirs at high water cut stages
  42. The collinear equilibrium points in the restricted three body problem with triaxial primaries
  43. Detection of the damage threshold of fused silica components and morphologies of repaired damage sites based on the beam deflection method
  44. On the bivariate spectral quasi-linearization method for solving the two-dimensional Bratu problem
  45. Ion acoustic quasi-soliton in an electron-positron-ion plasma with superthermal electrons and positrons
  46. Analysis of projectile motion in view of conformable derivative
  47. Computing multiple ABC index and multiple GA index of some grid graphs
  48. Terahertz pulse imaging: A novel denoising method by combing the ant colony algorithm with the compressive sensing
  49. Characteristics of microscopic pore-throat structure of tight oil reservoirs in Sichuan Basin measured by rate-controlled mercury injection
  50. An activity window model for social interaction structure on Twitter
  51. Transient thermal regime trough the constitutive matrix applied to asynchronous electrical machine using the cell method
  52. On the zagreb polynomials of benzenoid systems
  53. Integrability analysis of the partial differential equation describing the classical bond-pricing model of mathematical finance
  54. The Greek parameters of a continuous arithmetic Asian option pricing model via Laplace Adomian decomposition method
  55. Quantifying the global solar radiation received in Pietermaritzburg, KwaZulu-Natal to motivate the consumption of solar technologies
  56. Sturm-Liouville difference equations having Bessel and hydrogen atom potential type
  57. Study on the response characteristics of oil wells after deep profile control in low permeability fractured reservoirs
  58. Depiction and analysis of a modified theta shaped double negative metamaterial for satellite application
  59. An attempt to geometrize electromagnetism
  60. Structure of traveling wave solutions for some nonlinear models via modified mathematical method
  61. Thermo-convective instability in a rotating ferromagnetic fluid layer with temperature modulation
  62. Construction of new solitary wave solutions of generalized Zakharov-Kuznetsov-Benjamin-Bona-Mahony and simplified modified form of Camassa-Holm equations
  63. Effect of magnetic field and heat source on Upper-convected-maxwell fluid in a porous channel
  64. Physical cues of biomaterials guide stem cell fate of differentiation: The effect of elasticity of cell culture biomaterials
  65. Shooting method analysis in wire coating withdrawing from a bath of Oldroyd 8-constant fluid with temperature dependent viscosity
  66. Rank correlation between centrality metrics in complex networks: an empirical study
  67. Special Issue: The 18th International Symposium on Electromagnetic Fields in Mechatronics, Electrical and Electronic Engineering
  68. Modeling of electric and heat processes in spot resistance welding of cross-wire steel bars
  69. Dynamic characteristics of triaxial active control magnetic bearing with asymmetric structure
  70. Design optimization of an axial-field eddy-current magnetic coupling based on magneto-thermal analytical model
  71. Thermal constitutive matrix applied to asynchronous electrical machine using the cell method
  72. Temperature distribution around thin electroconductive layers created on composite textile substrates
  73. Model of the multipolar engine with decreased cogging torque by asymmetrical distribution of the magnets
  74. Analysis of spatial thermal field in a magnetic bearing
  75. Use of the mathematical model of the ignition system to analyze the spark discharge, including the destruction of spark plug electrodes
  76. Assessment of short/long term electric field strength measurements for a pilot district
  77. Simulation study and experimental results for detection and classification of the transient capacitor inrush current using discrete wavelet transform and artificial intelligence
  78. Magnetic transmission gear finite element simulation with iron pole hysteresis
  79. Pulsed excitation terahertz tomography – multiparametric approach
  80. Low and high frequency model of three phase transformer by frequency response analysis measurement
  81. Multivariable polynomial fitting of controlled single-phase nonlinear load of input current total harmonic distortion
  82. Optimal design of a for middle-low-speed maglev trains
  83. Eddy current modeling in linear and nonlinear multifilamentary composite materials
  84. The visual attention saliency map for movie retrospection
  85. AC/DC current ratio in a current superimposition variable flux reluctance machine
  86. Influence of material uncertainties on the RLC parameters of wound inductors modeled using the finite element method
  87. Cogging force reduction in linear tubular flux switching permanent-magnet machines
  88. Modeling hysteresis curves of La(FeCoSi)13 compound near the transition point with the GRUCAD model
  89. Electro-magneto-hydrodynamic lubrication
  90. 3-D Electromagnetic field analysis of wireless power transfer system using K computer
  91. Simplified simulation technique of rotating, induction heated, calender rolls for study of temperature field control
  92. Design, fabrication and testing of electroadhesive interdigital electrodes
  93. A method to reduce partial discharges in motor windings fed by PWM inverter
  94. Reluctance network lumped mechanical & thermal models for the modeling and predesign of concentrated flux synchronous machine
  95. Special Issue Applications of Nonlinear Dynamics
  96. Study on dynamic characteristics of silo-stock-foundation interaction system under seismic load
  97. Microblog topic evolution computing based on LDA algorithm
  98. Modeling the creep damage effect on the creep crack growth behavior of rotor steel
  99. Neighborhood condition for all fractional (g, f, n′, m)-critical deleted graphs
  100. Chinese open information extraction based on DBMCSS in the field of national information resources
  101. 10.1515/phys-2018-0079
  102. CPW-fed circularly-polarized antenna array with high front-to-back ratio and low-profile
  103. Intelligent Monitoring Network Construction based on the utilization of the Internet of things (IoT) in the Metallurgical Coking Process
  104. Temperature detection technology of power equipment based on Fiber Bragg Grating
  105. Research on a rotational speed control strategy of the mandrel in a rotary steering system
  106. Dynamic load balancing algorithm for large data flow in distributed complex networks
  107. Super-structured photonic crystal fiber Bragg grating biosensor image model based on sparse matrix
  108. Fractal-based techniques for physiological time series: An updated approach
  109. Analysis of the Imaging Characteristics of the KB and KBA X-ray Microscopes at Non-coaxial Grazing Incidence
  110. Application of modified culture Kalman filter in bearing fault diagnosis
  111. Exact solutions and conservation laws for the modified equal width-Burgers equation
  112. On topological properties of block shift and hierarchical hypercube networks
  113. Elastic properties and plane acoustic velocity of cubic Sr2CaMoO6 and Sr2CaWO6 from first-principles calculations
  114. A note on the transmission feasibility problem in networks
  115. Ontology learning algorithm using weak functions
  116. Diagnosis of the power frequency vacuum arc shape based on 2D-PIV
  117. Parametric simulation analysis and reliability of escalator truss
  118. A new algorithm for real economy benefit evaluation based on big data analysis
  119. Synergy analysis of agricultural economic cycle fluctuation based on ant colony algorithm
  120. Multi-level encryption algorithm for user-related information across social networks
  121. Multi-target tracking algorithm in intelligent transportation based on wireless sensor network
  122. Fast recognition method of moving video images based on BP neural networks
  123. Compressed sensing image restoration algorithm based on improved SURF operator
  124. Design of load optimal control algorithm for smart grid based on demand response in different scenarios
  125. Face recognition method based on GA-BP neural network algorithm
  126. Optimal path selection algorithm for mobile beacons in sensor network under non-dense distribution
  127. Localization and recognition algorithm for fuzzy anomaly data in big data networks
  128. Urban road traffic flow control under incidental congestion as a function of accident duration
  129. Optimization design of reconfiguration algorithm for high voltage power distribution network based on ant colony algorithm
  130. Feasibility simulation of aseismic structure design for long-span bridges
  131. Construction of renewable energy supply chain model based on LCA
  132. The tribological properties study of carbon fabric/ epoxy composites reinforced by nano-TiO2 and MWNTs
  133. A text-Image feature mapping algorithm based on transfer learning
  134. Fast recognition algorithm for static traffic sign information
  135. Topical Issue: Clean Energy: Materials, Processes and Energy Generation
  136. An investigation of the melting process of RT-35 filled circular thermal energy storage system
  137. Numerical analysis on the dynamic response of a plate-and-frame membrane humidifier for PEMFC vehicles under various operating conditions
  138. Energy converting layers for thin-film flexible photovoltaic structures
  139. Effect of convection heat transfer on thermal energy storage unit
Sign up now to receive a 20% welcome discount
Institutional Access
How does access work?
Have an idea on how to improve our website?
Please write us.
© 2025 De Gruyter Brill
Downloaded on 8.12.2025 from https://www.degruyterbrill.com/document/doi/10.1515/phys-2018-0027/html
Scroll to top button