Chair of
Multimedia Communications and Signal Processing
Prof. Dr.-Ing. André Kaup

Modeling the Processing Time of HEVC Decoding

Supervisor:Dipl.-Ing. Christian Herglotz (Room 01.178)
Faculty:Prof. Dr.-Ing. André Kaup
Student:Guer, Roman

For state-of-the-art video systems real-time capability is crucial. The current video coding standard HEVC achieves a much higher compression efficiency than its predecessors, but on the other hand its complexity has increased significantly. Thus it is of great importance to design processes in such a way that real-time capability is guaranteed.

To this end, one solution would be to encode the video in such a way that this criterion is met. As a consequence, the encoder has to know some properties of the decoder. When the encoder now knows that a specific decision results in a specific increase in decoding time, it can decide such that real-time decoding can be achieved.

These decoder characteristics can, e.g., be described by a model similar to the HEVC decoding energy model proposed earlier. In this work, it is the aim to find out if the model is also applicable to estimate the decoding time. To this end, two approaches can be tested: First, all coefficients can be linearly calculated using the results found by Elisabeth Walencik, second, they can be newly derived by corresponding measurements. Afterwards, the accuracy of the found model will be analyzed using real coded bit streams.

TypeResearch Internship