I'm interested in implementations of H.265/HEVC encoders, which can be used in practice. Since the H.265/HEVC standard only defines the decoder, finding good syntax elements is the job of the encoder. However, I guess finding the optimal syntax elements is infeasible, and thus, I assume that it is up to the designer of the encoder to develop Heuristics in order to determine good enough syntax elements within a reasonable time.
To get an idea how such Heuristics could look like, I had a look at two open source encoders:
x265 HEVC Encoder: https://bitbucket.org/multicoreware/x265_git/src/master/
However, I cannot find any document that is specifying either of both encoders. By just looking at the source code, it is very hard for me to understand how the Heuristics work.
Does by chance anyone know where I can find these specifications? And/or does anyone know a reasonable implementation of an HEVC encoder with a proper specification?
Thanks a lot!
[I had a look at the repositories jvet/HM (https://vcgit.hhi.fraunhofer.de/jvet/HM) and x265 HEVC Encoder (https://bitbucket.org/multicoreware/x265_git/src/master/) but could not find a document that is specifying the used algorithms. I tried to find other open source encoders but was not successful.]
