VMAF Re-implementation on PyTorch: Some Experimental ResultsKirill Aistov, Maxim Koroteev
Based on the standard VMAF implementation we propose an implementation of VMAF using PyTorch framework. For this implementation comparisons with the standard (libvmaf) show the discrepancy $\lesssim 10^{-2}$ in VMAF units. We investigate gradients computation when using VMAF as an objective function and demonstrate that training using this function does not result in ill-behaving gradients. The implementation is then used to train a preprocessing filter. It is demonstrated that its performance is superior to the unsharp masking filter. The resulting filter is also easy for implementation and can be applied in video processing tasks for video copression improvement. This is confirmed by the results of numerical experiments.
1.4CVDec 20, 2022
Image quality prediction using synthetic and natural codebooks: comparative resultsMaxim Koroteev, Kirill Aistov, Valeriy Berezovskiy et al.
We investigate a model for image/video quality assessment based on building a set of codevectors representing in a sense some basic properties of images, similar to well-known CORNIA model. We analyze the codebook building method and propose some modifications for it. Also the algorithm is investigated from the point of inference time reduction. Both natural and synthetic images are used for building codebooks and some analysis of synthetic images used for codebooks is provided. It is demonstrated the results on quality assessment may be improves with the use if synthetic images for codebook construction. We also demonstrate regimes of the algorithm in which real time execution on CPU is possible for sufficiently high correlations with mean opinion score (MOS). Various pooling strategies are considered as well as the problem of metric sensitivity to bitrate.
2.6CVJan 27, 2022
On scale-invariant properties in natural images and their simulationsMaxim Koroteev, Kirill Aistov
We study samples of natural images for which a set of statistical characteristics is computed and scale-invariant properties of samples are demonstrated computationally. Computations of the power spectrum are carried out and a power-law decaying power spectrum is observed on samples taken from van Hateren images of natural scenes. We propose a dynamic model to reproduce the observed slope in the power spectrum qualitatively. For two types of sources for this model the behaviour of power spectrum is investigated and scale-invariance confirmed numerically. We then discuss potential applications of scale-invariant properties of natural images.