Attentive mechanisms for dynamic and static scene analysis

Milanese, Ruggero; Gil, Sylvia; Pun, Thierry

doi:10.1117/12.205668

Cited by 68 publications

(43 citation statements)

References 2 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…It is well known that motion is a strong indicator of saliency in dynamic visual scenes [36], [63], [67], so it is not surprising that MVs would be a powerful cue for saliency estimation. PMES estimates saliency by considering two properties: large motion magnitude in a spatio-temporal region, and the lack of coherence among MV angles in that region.…”

Section: Discussionmentioning

confidence: 99%

Compressed-domain visual saliency models: a comparative study

Khatoonabadi

Bajić

Shan

2016

Multimed Tools Appl

View full text Add to dashboard Cite

Abstract-Computational modeling of visual saliency has become an important research problem in recent years, with applications in video quality estimation, video compression, object tracking, retargeting, summarization, and so on. While most visual saliency models for dynamic scenes operate on raw video, several models have been developed for use with compresseddomain information such as motion vectors and transform coefficients. This paper presents a comparative study of eleven such models as well as two high-performing pixel-domain saliency models on two eye-tracking datasets using several comparison metrics. The results indicate that highly accurate saliency estimation is possible based only on a partially decoded video bitstream. The strategies that have shown success in compressed-domain saliency modeling are highlighted, and certain challenges are identified as potential avenues for further improvement.

show abstract

Section: Discussionmentioning

confidence: 99%

Compressed-domain visual saliency models: a comparative study

Khatoonabadi

Bajić

Shan

2016

Multimed Tools Appl

View full text Add to dashboard Cite

show abstract

“…The hybrid model is an adaptive motion-time dynamic hybrid model inspired by the motion-prior idea given by [13]. This model is based on the fact of the visual character, to combine the dynamic saliency maps and static saliency maps.…”

Section: Hybrid Modelmentioning

confidence: 99%

A New Hybrid Model for Video Shot Saliency Extraction

Fang¹

2014

Proceedings of the 3rd International Conference on Computer Science and Service System

View full text Add to dashboard Cite

Abstrac--Video screen shots of images are special, and they are motive compare to other images. To extract saliency maps from video images quickly and accurately is a hard task. This paper is inspired by some excellent works, employing the combination of several ideas to propose a hybrid model for extracting saliency. The hybrid model to extract salient region is based on visual attention theory, dynamic feature saliency maps extraction technique, and motionprior idea integrate spatial with time. We perform experiments to make extraction with itti model, and results prove the fact that the proposed model can get an ideal effect when extract salient region dynamically considering the combination of speed and direction is closing to biological visual mechanisms.

show abstract

“…The early work of Milanese et al (1995), one of the first computational visual saliency models, was based on such a competition that was implemented in the model as a series of continuous inter-and intrapixel differences. Toward this direction we implement competition by a constrained minimization approach that involves interscale, intrafeature, and interfeature constraints ).…”

Section: Volumetric Saliency By Feature Competitionmentioning

confidence: 99%

“…One of the most widely exploited computational models is the one introduced by Itti et al (1998), which is based on the model of Koch and Ullman and hypothesize that various visual features feed into a unique saliency map (Koch and Ullman 1985) that encodes the importance of each minor visual unit. Among the wealth of methods based on saliency maps, the differences are mainly due to the selection of feature maps to be included in the model [contrast (May and Zhang 2003), color (Itti et al 1998), orientation (Itti et al 1998), edges (Park et al 2002), size, skin maps (Rapantzikos and Tsapatsoulis 2005), texture, motion (Rapantzikos and Tsapatsoulis 2005;Milanese et al 1995;Abrams and Christ 2003), etc. ], the linear, or nonlinear operations [multiscale center-surround (Itti et al 1998), entropy (Kadir and Brady 2001), etc.…”

Section: Bottom-up Visual Attentionmentioning

confidence: 99%

Vision, Attention Control, and Goals Creation System

Rapantzikos

Avrithis²,

Kolias³

2010

Perception-Action Cycle

View full text Add to dashboard Cite

Biological visual attention has been long studied by experts in the field of cognitive psychology. The Holy Grail of this study is the exact modeling of the interaction between the visual sensory and the process of perception. It seems that there is an informal agreement on the four important functions of the attention process: (a) the bottom-up process, which is responsible for the saliency of the input stimuli; (b) the top-down process that bias attention toward known areas or regions of predefined characteristics; (c) the attentional selection that fuses information derived from the two previous processes and enables focus; and (d) the dynamic evolution of the attentional selection process. In the following, we will outline established computational solutions for each of the four functions. OverviewMost of our impressions and memories are based on vision. Nevertheless vision mechanisms and functionalities are still not apparent. How do we perceive shape, color, or motion and how do we automatically focus on the most informative parts of the visual input? It has been long established that primates, including human, use focused attention and fast saccades to analyze visual stimuli based on the current situation or the desired goal. Neuroscientists have proven that neural information related to shape, motion, and color is transmitted through, at least, three parallel and interconnected channels to the brain rather than a single one. Hence a second question arises related to how these channels are "linked" in order to provide useful information to the brain.The Human Visual System (HVS) creates a perceptual representation of the world that is quite different than the two dimensional depiction of the retina.K. Rapantzikos ( ) Image, Video and

show abstract

Attentive mechanisms for dynamic and static scene analysis

Cited by 68 publications

References 2 publications

Compressed-domain visual saliency models: a comparative study

Compressed-domain visual saliency models: a comparative study

A New Hybrid Model for Video Shot Saliency Extraction

Vision, Attention Control, and Goals Creation System

Contact Info

Product

Resources

About