Investigations on the post-processing headroom in lossy comperessed professional HD video formats

OData support
Mócsai Tamás
Department of Networked Systems and Services

With the continous appearance of new standards and service sin video technology, it is also necessary to keep up the simultaneous improvment of the technical devices..For obvious reasons the uncompressed video contains plenty of data, of which if we want to store or transmit it is crucial to compress without damaging the quailty of the pictures.

Ideally the post-processing occurs on the uncompressed video. However if we consider the aspects of storage in practice we may need to apply post-processing on compressed video as well. The codecs, which perform the compression of the video, are based on such psychovisual principles, which cause none or minimal subjective quality loss. The question is if the post-processing of compressed video increases the visibility or the significance of the codec quality loss.

The topic of my thesis is investigations on the post-processing headroom in lossy compressed professional HD video formats. Most of the inspected professional video compressions provide transparent quality through multi-generation coding/decoding. My task is to evaluate if these codecs/bitrates provide enough data headroom for certain post-processing steps (such as color correction). It is essential to know if these post-processing steps could be done properly or these steps just make the effect of the deteriorating quality of the lossy codec worse.

The thesis summarize the theoretical knowledge needed for the observation, starting with compression methods of motion picture. Then I continue reviewing the attributes of the used codecs (Apple ProRes, Sony XDCAM, Avid DNxHD and AVC-Intra) and objective quality metrics applied for analizing. I demonstrate the performed post-procession steps and finally with the help of a test environment I compare and investigate the received results.


Please sign in to download the files of this thesis.