No abstract

This paper addresses the problem of quality estimation of digitally coded video sequences. The topic is of great interest since many products in digital video are about to be released and it is thus important t o h a v e robust methodologies for testing and performance evaluation of such devices. The inherent problem is that human vision has to be taken into account in order to assess the quality of a sequence with a good correlation with human judgment. It is well known that the commonly used metric, the signal-to-noise ratio is not correlated with human vision.A metric for the assessment of video coding quality is presented. It is based on a multi-channel model of human spatio-temporal vision that has been parameterized for video coding applications by psychophysical experiments. The visual mechanisms of vision are simulated by a spatio-temporal lter bank. The decomposition is then used to account for phenomena as contrast sensitivity and masking. Once the amount of distortions actually perceived is known, quality estimation can be assessed at various levels. The described metric is able to rate the overall quality of the decoded video sequence as well as the rendition of important features of the sequence such a s contours or textures.

No abstract

No abstract

One solution to the lack of label problem is to exploit transfer learning, whereby one acquires knowledge from source-domains to improve the learning performance in the target-domain. The main challenge is that the source and target domains may have different distributions. An open problem is how to select the available models (including algorithms and parameters) and importantly, abundance of source-domain data, through statistically reliable methods, thus making transfer learning practical and easy-to-use for real-world applications. To address this challenge, one needs to take into account the difference in both marginal and conditional distributions in the same time, but not just one of them. In this paper, we formulate a new criterion to overcome "double" distribution shift and present a practical approach "Transfer Cross Validation" (TrCV) to select both models and data in a cross validation framework, optimized for transfer learning. The idea is to use density ratio weighting to overcome the difference in marginal distributions and propose a "reverse validation" procedure to quantify how well a model approximates the true conditional distribution of target-domain. The usefulness of TrCV is demonstrated on different cross-domain tasks, including wine quality evaluation, web-user ranking and text categorization. The experiment results show that the proposed method outperforms both traditional cross-validation and one state-of-the-art method which only considers marginal distribution shift. The software and datasets are available from the authors.