On Sat, Sep 21, 2013 at 04:47:30PM +0200, Vittorio Giovara wrote: > On Mon, Sep 16, 2013 at 11:35 AM, Luca Barbato <[email protected]> wrote: > > On 22/01/13 20:30, Jordi Ortiz wrote: > >> Thanks for the review > >> > >>> What is the status of this decoder now? Does it still produce artifacts? > >>> If yes, for which cases? > >> > >> The status is the same as it was. When sub pixel motion vector precision > >> is used the decoded result isn't bit exact with last libschroedinger > >> version. > >> Also deep-estimation is not working properly. If anyone knows what > >> deep-estimation is and wants to point it out it would be very welcome, I > >> didn't manage to find any reference in the standard. I know that deep > >> estimation uses mv_precision values over zero so the problem in this > >> case could be related to the mv_precision. > > > > Let's just document that and add some avpriv_report_missing_feature. > > > > Would be nice having it in some shape this year. > > I do agree, Jordi think that you could send an updated patch with the > latest changes?
My guess would be that this needs some improved meaningful return values and avpriv_report_missing_feature somewhere on top of the rebase. Diego _______________________________________________ libav-devel mailing list [email protected] https://lists.libav.org/mailman/listinfo/libav-devel
