Objective and subjective assessment of space-transcoded videos for mobile receivers

In a digital television scenario, the video signal may have different bit rates, encoding formats, and spatial resolutions according to the type of transmission, the application, and the receiver. To flexibly deliver video to users with different available resources, the content needs to be adapted dynamically. Video transcoding is the operation of converting a video from one format into another. In the particular case of the reception of digital videos using mobile receivers, the spatial resolution of the video must be adjusted in order to fit in the display of a mobile phone. In this work, we present an analysis of the subjective and objective quality of spatially transcoded videos. We consider transcoding algorithms that down sample the video frames using moving average, median, mode, weighted average and sigma filters with different sizes. The results show that among the tested algorithms, the spatial transcoded algorithms using the filters 2 times 2 median and 2 times 2 sigma filters produce the best results, both objectively and subjectively.