Modelling visual attention and motion effect for visual quality evaluation
0
Citation
21
Reference
10
Related Paper
Abstract:
The perceptual visual quality evaluation of Human Visual System (HVS) is very complex. It concerns almost all aspects of visual processing in vision path, from low-level neuron activities to high-level visual perception. Existing perceptual Visual Quality Metrics (VQMs) only considered several of the mechanisms of HVS and many others are ignored. In this paper, two global modulatory factors, visual attention and motion suppression, are modelled and combined to form a mathematic expression - Perceptual Quality Significant Level (PQSL). To a certain extent, it is believed that PQSL value reflect the processing ability of human brain on local visual contents. To evaluate their effects on visual quality evaluation, two VQMs are proposed. One is a MSE-like VQM based on PQSL-modulated JND profile, which was proposed in Z.K. Lu et al., (2004); the other VQM is based on Wang's visual quality assessment Zhou Wang et al., (2004) PQSL values are used to adjust the weights of his structural similarity index. Experimental results show that introducing of the global modulatory factors can improve the performance of current visual quality metrics.Keywords:
Visual processing
Similarity (geometry)
The perceptual visual quality evaluation of human visual system (HVS) is very complex. It concerns almost all aspects of visual processing in the vision path, from low-level neuron activities to high-level visual perception. Existing perceptual visual quality metrics (VQMs) only considered several of the mechanisms of HVS, and many others are ignored. In this paper, two global modulatory factors, visual attention and motion suppression, are modelled and combined to form a mathematic expression - perceptual quality significant level (PQSL). To a certain extent, it is believed that PQSL values reflect the processing ability of the human brain on local visual content. To evaluate their effects on visual quality evaluation, two VQMs are proposed. One is a MSE-like VQM based in a PQSL-modulated JND profile, which was proposed in (Z. K. Lu et al, Proc. ICASSP'2004, v.3, p.705-708, 2004); the other VQM is based on Wang's visual quality assessment (Sig. Proc.:Image Comm., v.19, n.2, p.121-132, 2004), PQSL values are used to adjust the weights of his structural similarity index. Experimental results show that introduction of the global modulatory factors can improve the performance of current visual quality metrics.
Visual processing
Similarity (geometry)
Cite
Citations (1)
The processing power of the human visual system is truly awe-inspiring. Hundreds of millions of years of natural selection have led to the evolution of a visual system whose rapidity and accuracy is quite breathtaking. Images flashed on a screen can often be identified on the basis of only 100 to 150 ms of processing (see Thorpe & Imbert, 1989). Furthermore, the number of different objects and scenes that can be identified by the average human is probably well over 100,000. Such levels of performance make the processing capacities of current image processing technology look positively feeble. And the contrast between biological and artificial vision is made even more striking when one takes into account the fact that the components used by biological visual systems, namely neurons, are several orders of magnitude slower than the components used in current electronics: neurons typically operate with a maximum pulse rate of around 100 Hz, whereas even the humble PC uses components that operate with a clock rate of 10 MHz or more.
Visual processing
Cite
Citations (2)
Current state-of-the-art algorithms that process visual information for end use by humans treat images and video as traditional "signals" and employ sophisticated signal processing strategies to achieve their excellent performance. These algorithms also incorporate characteristics of the human visual system (HVS), but typically in a relatively simplistic manner, and achievable performance is reaching an asymptote. However, large gains are still realizable with current techniques by aggressively incorporating HVS characteristics to a much greater extent than is presently done. Achieving these gains requires HVS characterizations which better model natural image perception ranging from sub-threshold perception (where distortions are not visible) to supra-threshold perception (where distortions are clearly visible). This paper reviews classical psychophysical HVS characterizations focused on the visual cortex (V1), pertaining to the contrast sensitivity function, summation, and masking, which have been obtained using unrealistic stimuli such as sinusoids and white noise. The direct applicability of these results to natural images is often not clear. Complementary results are then presented which have been obtained using realistic stimuli derived from or consisting of natural images, along with several applications of these results. Finally, a new structure-based masking model is proposed to model masking in homogeneous natural image patches as a function of the image type: textures, edges, or structures.
Visual masking
Filling-in
Visual processing
Asymptote
Cite
Citations (7)
The perceptual visual quality evaluation of Human Visual System (HVS) is very complex. It concerns almost all aspects of visual processing in vision path, from low-level neuron activities to high-level visual perception. Existing perceptual Visual Quality Metrics (VQMs) only considered several of the mechanisms of HVS and many others are ignored. In this paper, two global modulatory factors, visual attention and motion suppression, are modelled and combined to form a mathematic expression - Perceptual Quality Significant Level (PQSL). To a certain extent, it is believed that PQSL value reflect the processing ability of human brain on local visual contents. To evaluate their effects on visual quality evaluation, two VQMs are proposed. One is a MSE-like VQM based on PQSL-modulated JND profile, which was proposed in Z.K. Lu et al., (2004); the other VQM is based on Wang's visual quality assessment Zhou Wang et al., (2004) PQSL values are used to adjust the weights of his structural similarity index. Experimental results show that introducing of the global modulatory factors can improve the performance of current visual quality metrics.
Visual processing
Similarity (geometry)
Cite
Citations (0)
Evidence reveals that visual processing speed decreases with age. The Motor-Free Visual Perception Test- Third Edition (MVPT-3) has an age-normed Response Time Index that measures visual processing speed. In 2015, a new version, Motor-Free Visual Perception Test- Fourth Edition (MVPT-4), was published. The new MVPT-4 does not yet demonstrate its utility in measuring visual processing speed. The purpose of this study was to explore if differences in visual processing speed between younger adults ages 20-35 years and older adults ages 70 years and older could be detected using the new MVPT-4. Results revealed a significant difference between older and younger adults’ time to complete the MVPT-4 (p <.05). This pilot study demonstrated that the MVPT-4 may be able to detect age-related changes in visual processing speed and therefore, a possible clinical tool for occupational therapists.
Visual processing
Cite
Citations (1)
Perceptual quality metrics are widely deployed in image and video processing systems. These metrics aim to emulate the integral mechanisms of the human visual system (HVS) to correlate well with visual perception of quality. One integral property of the HVS is, however, often neglected: visual attention (VA) [1]. The essential mechanisms associated with VA consist mainly of higher cognitive processing, deployed to reduce the complexity of scene analysis. For this purpose, a subset of the visual information is selected by shifting the focus of attention across the visual scene to the most relevant objects. By neglecting VA, perceptual quality models inherently assume that all objects draw the attention of the viewer to the same degree.
Visual processing
Cite
Citations (129)
Scene statistics
Visual processing
Visual Search
Cite
Citations (32)
Abstract The human visual system is a remarkably versatile and powerful spatial image processor. The eye itself is only a small part of the visual mechanism. The task of analyzing and comprehending the visual world is so complex that the brain devotes between one‐third and one‐half of the entire cerebral cortex to solving it: all of the occipital lobe and large parts of the parietal, temporal, and frontal lobes as well. To appreciate the magnitude of the task of visual processing, it is useful to consider some of the problems that the visual system must address in the course of analyzing visual images. Solutions to initial problems are discussed. Details on the visual processing of the eye including visual analysis and coding strategies and cortical processing are given.
Visual processing
Gaze-contingency paradigm
Occipital lobe
N2pc
Parietal lobe
Cite
Citations (0)
Visual processing
Cite
Citations (16)
Older adults are at a significantly increased risk of being involved in motor vehicle accidents. Evidence reveals that visual processing speed decreases with age, which may impact driving. The Motor-Free Visual Perception Test- Third Edition (MVPT-3) is used as a pre-driving assessment and has an age-normed Response Time Index that measures visual processing speed. In 2015, a new version, the new Motor-Free Visual Perception Test- Fourth Edition (MVPT-4), was published. The new MVPT-4 does not yet demonstrate its utility in measuring visual processing speed. The purpose of this study was to explore if differences in visual processing speed between younger adults ages 20-35 years and older adults ages 70 years and older could be detected using the new MVPT-4. Results revealed a significant difference between older and younger adults’ time to complete the MVPT-4 (p
Visual processing
Cite
Citations (0)