No-reference algorithms for video quality assessment based on artifact evaluation in MPEG-2 and H.264 encoding standards

摘要

In this paper, we propose two algorithms to assess quality in video sequences when no reference is available. The algorithms are the result of measuring artifacts such as blurring or tiling to the decoded image; firstly tested in MPEG-2 compression standard, and subsequently, analyzing the passage from this standard to its evolution H.264, taking into account its advanced techniques to mask the degradation, for example, the deblocking filtering or the variation in macroblocks size. The result of each metric affects in a different way depending on the spatial and temporal complexity of the video sequence. Several tests have been developed in a collection of representative sequences to demonstrate the efficiency of the algorithms.