Person: Narwaria, Manish
Loading...
Name
Manish Narwaria
Job Title
Email Address
Telephone
Birth Date
Specialization
6 results
Search Results
Now showing 1 - 6 of 6
Publication Metadata only No-reference video quality measurement: Added value of machine learning(SPIE, 01-12-2015) Mocanu, DC; Pokhrel, Jeevan; Garella, Juan Pablo; Sepp'nen, Janne; Liotou, Eirini; Narwaria, Manish; DA-IICT, GandhinagarVideo quality measurement is an important component in the end-to-end video delivery chain. Video quality is, however, subjective and thus there will always be inter-observer differences in the subjective opinion about the visual quality of the same video. Despite this, most existing works on objective quality measurement typically focus only on predicting a single score, and evaluate their prediction accuracies based on how close it is to the mean opinion scores (or similar average based ratings). Clearly, such an approach ignores the underlying diversities in the subjective scoring process, and as a result, does not allow further analysis on how reliable the objective prediction is in terms of subjective variability. Consequently, the aim of this paper is to analyze this issue and present a machine learning based solution to address it. We demonstrate the utility of our ideas by considering the practical scenario of video broadcast transmissions with focus on Digital Terrestrial Television (DTT), and proposing a no-reference objective video quality estimator for such application. We conducted meaningful verification studies on different video content (including video clips recorded from real DTT Broadcast transmissions) in order to verify the performance of the proposed solution.Publication Metadata only Toward Better Statistical Validation of Machine Learning-Based Multimedia Quality Estimators(IEEE, 01-06-2018) Narwaria, Manish; DA-IICT, GandhinagarObjective assessment of multimedia quality using machine learning (ML) has been gaining popularity especially in the context of both traditional (e.g., terrestrial and satellite broadcast) and advance (such as over-the-top media services, IPTV) broadcast services. Being data-driven, these methods obviously rely on training to find the optimal model parameters. Therefore, to statistically compare and validate such ML-based quality predictors, the current approach randomly splits the given data into training and test sets and obtains a performance measure (for instance mean squared error, correlation coefficient etc.). The process is repeated a large number of times and parametric tests (e.g., t test) are then employed to statistically compare mean (or median) prediction accuracies. However, the current approach suffers from a few limitations (related to the qualitative aspects of training and testing data, the use of improper sample size for statistical testing, possibly dependent sample observations, and a lack of focus on quantifying the learning ability of the ML-based objective quality predictor) which have not been addressed in literature. Therefore, the main goal of this paper is to shed light on the said limitations both from practical and theoretical perspectives wherever applicable, and in the process propose an alternate approach to overcome some of them. As a major advantage, the proposed guidelines not only help in a theoretically more grounded statistical comparison but also provide useful insights into how well the ML-based objective quality predictors exploit data structure for learning. We demonstrate the added value of the proposed set of guidelines on standard datasets by comparing the performance of few existing ML-based quality estimators. A software implementation of the presented guidelines is also made publicly available to enable researchers and developers to test and compare different models in a repeatable manner.Publication Metadata only Data Analysis in Multimedia Quality Assessment: Revisiting the Statistical Tests(IEEE, 01-08-2018) Narwaria, Manish; Krasula, Lukas; Callet, Patrick Le; DA-IICT, GandhinagarAssessment of multimedia quality relies heavily on subjective assessment, and is typically done by human subjects in the form of preferences or continuous ratings. Such data are crucial for analysis of different multimedia-processing algorithms as well as validation of objective (computational) methods for the said purpose. To that end, statistical testing provides a theoretical framework toward drawing meaningful inferences, and making well-grounded conclusions and recommendations. While parametric tests (such as t test, ANOVA, and error estimates like confidence intervals) are popular and widely used in the community, there appears to be a certain degree of confusion in the application of such tests. Specifically, the assumptions of normality and homogeneity of variance are often not well understood, leading to incorrect application and/or interpretation of the statistical test results. Therefore, the main goal of this paper is to present new guidelines toward proper use of statistical tests and, hence, fix some of the issues in multimedia quality assessment. The said guidelines are derived based on theoretical analysis of sampling distribution of test statistics, and consider practical aspects of data analysis in the said domain. Experimental results on both simulated and real data are presented to support the arguments made. Software that implements the said recommendations is also made publicly available, in order to help researchers and practitioners perform correct statistical comparison of models.Publication Metadata only Preference of Experience in Image Tone-Mapping: Dataset and Framework for Objective Measures Comparison(IEEE, 01-12-2017) Krasula, Lukas; Narwaria, Manish; Fliegel, Karel; Callet, Patrick Le; DA-IICT, GandhinagarThe popularity of high dynamic range (HDR) imaging has grown in both academic and private research sectors. Since the native visualization of HDR content still has its limitations, the importance of dynamic range compression (i.e., tone-mapping) is very high. This paper evaluates observers' preference of experience in context of image tone-mapping. Given the different nature of natural and computer-generated content, the way observers perceive the quality of tone-mapped images can be fundamentally different. In this paper, we describe a subjective experiment attempting to determine users' preference with respect to these two types of content in two different viewing scenarios-with and without the HDR reference. The results show that the absence of the reference can significantly influence the subjects' preferences for the natural images, while no significant impact has been found in the case of the synthetic images. Moreover, we introduce a benchmarking framework and compare the performance of selected objective metrics. The resulting dataset and framework are made publicly available to provide a common test bed and methodology for evaluating metrics in the considered scenario.Publication Metadata only Lorentzian Based Adaptive Filters for Impulsive Noise Environments(IEEE, 01-02-2017) Das, Rajib Lochan; Narwaria, Manish; DA-IICT, GandhinagarIn this paper, three Lorentzian based robust adaptive algorithms are proposed for identifying systems in presence of impulsive noise. The first algorithm called Lorentzian adaptive filtering (LAF) is derived from a sliding window type cost function with Lorentzian norm of past errors to combat adverse effect of impulsive noise on systems. The first and second order convergence analyses of the LAF algorithm are carried out in this paper. Then, to identify sparse systems in impulsive noise environment, l0�norm penalty is introduced to the cost function of the LAF algorithm leading to a new algorithm called Lorentzian hard thresholding adaptive filtering (LHTAF) which employs hard thresholding operator with a fixed hard thresholding parameter to obtain sparse solutions. The effect of the hard thresholding operator is further analyzed, and the analysis shows that a variable hard thresholding parameter offers significant improvement in the performance of the algorithm, and this result in the final algorithm called Lorentzian variable hard thresholding adaptive filtering (LVHTAF) where the hard thresholding parameter is adjusted adaptively. Simulation results show that the LVTHAF outperforms the existing robust sparse adaptive algorithms by producing lesser steady state mean square error.Publication Metadata only Interval-Based Least Squares for Uncertainty-Aware Learning in Human-Centric Multimedia Systems(IEEE, 11-11-2021) Narwaria, Manish; Tatu, Aditya; DA-IICT, GandhinagarMachine learning (ML) methods are popular in several application areas of multimedia signal processing. However, most existing solutions in the said area, including the popular least squares, rely on penalizing predictions that deviate from the target ground-truth values. In other words, uncertainty in the ground-truth data is simply ignored. As a result, optimization and validation overemphasize a single-target value when, in fact, human subjects themselves did not unanimously agree to it. This leads to an unreasonable scenario where the trained model is not allowed the benefit of the doubt in terms of prediction accuracy. The problem becomes even more significant in the context of more recent human-centric and immersive multimedia systems where user feedback and interaction are influenced by higher degrees of freedom (leading to higher levels of uncertainty in the ground truth). To ameliorate this drawback, we propose an uncertainty aware loss function (referred to as�MSE?�) that explicitly accounts for data uncertainty and is useful for both optimization (training) and validation. As examples, we demonstrate the utility of the proposed method for blind estimation of perceptual quality of audiovisual signals, panoramic images, and images affected by camera-induced distortions. The experimental results support the theoretical ideas in terms of reducing prediction errors. The proposed method is also relevant in the context of more recent paradigms, such as crowdsourcing, where larger uncertainty in ground truth is expected.