[Libav-user] Problem with the writing of video

classic Classic list List threaded Threaded
3 messages Options
Reply | Threaded
Open this post in threaded view
|

[Libav-user] Problem with the writing of video

Pierre Delmas
Hello,


I have written yesterday for a problem that I have solved. But now I
have another problem I'm able to write headers and almost to write
frames (as you can think its the almost the problem). Currently I read a
video and try to re-write at the same but when I do this I have a video
who is generated with a size less (it's a third of the original
approximatively) than the original video and a duration of zero seconds.
I write my code in C++


At reading I push the pointers of extracted AVFrame into a vector<AVFrame*>


You can find attached my code for the writing part and a text document
who contains the execution.


It seems that only I and P frames are written but I can certify that
this video contains also B-frames.


Is there a better way to extract AVFrame than use a vector ? If yes do
you have examples ?

Is there any other (and probably better) way to write a video ?


Thank you for your help,

Best regards,

DELMAS P.



_______________________________________________
Libav-user mailing list
[hidden email]
http://ffmpeg.org/mailman/listinfo/libav-user

signature.asc (484 bytes) Download Attachment
Reply | Threaded
Open this post in threaded view
|

Re: Problem with the writing of video

Pierre Delmas

And I forgot the attached documents my bad sorry.


Thank you again for your help,

Best regards,

DELMAS P.


Le 16/04/2017 à 13:15, Pierre Delmas a écrit :
Hello,


I have written yesterday for a problem that I have solved. But now I
have another problem I'm able to write headers and almost to write
frames (as you can think its the almost the problem). Currently I read a
video and try to re-write at the same but when I do this I have a video
who is generated with a size less (it's a third of the original
approximatively) than the original video and a duration of zero seconds.
I write my code in C++


At reading I push the pointers of extracted AVFrame into a vector<AVFrame*>


You can find attached my code for the writing part and a text document
who contains the execution.


It seems that only I and P frames are written but I can certify that
this video contains also B-frames.


Is there a better way to extract AVFrame than use a vector ? If yes do
you have examples ?

Is there any other (and probably better) way to write a video ?


Thank you for your help,

Best regards,

DELMAS P.




_______________________________________________
Libav-user mailing list
[hidden email]
http://ffmpeg.org/mailman/listinfo/libav-user


_______________________________________________
Libav-user mailing list
[hidden email]
http://ffmpeg.org/mailman/listinfo/libav-user

datas.txt (57K) Download Attachment
video.cpp (2K) Download Attachment
signature.asc (484 bytes) Download Attachment
Reply | Threaded
Open this post in threaded view
|

Re: Problem with the writing of video

Pierre Delmas
Hello,


I have made some progress until my last mail 3 hours ago. Now I write
the video but all my frames are on each others. It's like the video is
very very accelerated I don't understand why ?. So here is my code if
anyone has an idea.


Thank you for your help,

Best regards,

DELMAS P.


Le 16/04/2017 à 13:17, Pierre Delmas a écrit :
> non-strictly-monotonic PTS


_______________________________________________
Libav-user mailing list
[hidden email]
http://ffmpeg.org/mailman/listinfo/libav-user

video.cpp (3K) Download Attachment
signature.asc (484 bytes) Download Attachment