Hossein Malekmohamadi

Perceptual Video Quality Metric for 3D video quality assessment

One method of evaluating the quality of stereoscopic video is the use of conventional two dimensional (2D) objective metrics. Metrics with good representation of the Human Visual System (HVS) will present more accurate evaluation. In this paper we propose a perceptual based objective metric for 2D videos for 3D video quality evaluation. The proposed Perceptual Quality Metric (PQM) shows better results for 3D video quality evaluation and outperforms the Video Quality Metric (VQM); as it is sensitive to slight changes in image degradation and error quantification starts at pixel level right up to the sequence level. Verifications are done through series of subjective tests to show the level of correlation of PQM and user scores.

A new reduced reference metric for color plus depth 3D video

A new reduced reference (RR) objective quality metric for 3D video is proposed that incorporates spatial neighboring information. The contrast measures from gray level co-occurrence matrices (GLCM) for both color and depth sections are main parts of spatial information. Side information is extracted from edge properties of reference 3D video and sent through an auxiliary channel. The other important factor in the proposed metric is the unequal weight of color and depth sections, which can maximize the performance of the proposed metric for some specific values. Performance of the proposed metric is validated through series of subjective tests. For validations, compression and transmission artifacts are considered. The average correlation of the proposed metric and subjective quality scores is 0.82 for compressed 3D videos when color to depth importance ratio is near 0.8. This measure for transmitted 3D videos is 0.857 for the same value of color to depth importance ratio.

Automatic QOE Prediction in Stereoscopic Videos

In this paper, we propose a method for automatic quality of experience (QoE) prediction in stereoscopic videos. QoE, though embodying the subjective measures of the end users perceived quality, can be expressed in relation to some quality of service (QoS) parameters. Having information on content types in modelling QoE-QoS interactions is advantageous as videos with the same QoS parameters may have different subjective scores due to different content types. Consequently, using content clustering with the help of spatio-temporal activities within depth layers, QoE predictor is designed per each content cluster utilising full reference (FR) and no reference (NR) metrics. Finally, the performance of the proposed QoE prediction algorithm is evaluated extensively and the overall measure of success value equal to 95.4% is achieved for the test sequences. This model can be applied for QoE control in video provisioning systems.

Subjective quality estimation based on neural networks for stereoscopic videos

A neural network based technique is proposed to estimate subjective quality of stereoscopic videos. Moreover, to utilize this model for applications where availability of reference signal is not possible to receiver, it applies objective quality of video with minimum dependency on reference signal. This paper presents fast, accurate and consistent subjective quality estimation. Feasibility and accuracy of the proposed technique is thoroughly analyzed with extensive subjective experiments and simulations. Results illustrate that performance measure of 92.3% in subjective quality estimation can be achieved with the proposed technique.

A new reduced reference objective quality metric for stereoscopic video

A new reduced reference (RR) objective quality metric for stereoscopic video is proposed that incorporates spatial neighbouring information. The contrast measures from grey level co-occurrence matrices (GLCM) for both colour and depth are the main parts of spatial information. Side information are extracted from edge information of original video and sent through an auxiliary channel. The other important factor in the proposed metric is the unequal weight of colour and depth views, which can maximise the performance of the proposed metric for some specific values. Performance of the proposed metric is validated through series of subjective tests to show how it correlates subjective quality scores. The average correlation of the proposed metric and subjective quality scores is 0.82 when colour to depth importance ratio is near 4.