I wanted to use ffmpeg with the CUDA video encoder/decoder API since it would have saved a lot of my time but it seems that making the 2 run together would be more of a waste of time then saving. So I have 2 very specific question regarding that:
Since the CUDA Encoder API reads YUV format, encodes into h.264 and writes back into .264 container. Now I want to read .VOB file into YUV, do I use the CUDA decoder for that? And also, once I have the encoded .264 file, how do I write it into a MP4 container?
How are you guys managing the Audio part of the video? Are there any best practices or tutorial I can look at?
I wanted to use ffmpeg with the CUDA video encoder/decoder API since it would have saved a lot of my time but it seems that making the 2 run together would be more of a waste of time then saving. So I have 2 very specific question regarding that:
Since the CUDA Encoder API reads YUV format, encodes into h.264 and writes back into .264 container. Now I want to read .VOB file into YUV, do I use the CUDA decoder for that? And also, once I have the encoded .264 file, how do I write it into a MP4 container?
How are you guys managing the Audio part of the video? Are there any best practices or tutorial I can look at?
for a start, look for the commandline tool called mp4box.exe. You can find it on the net. You can use this to convert the files generated with the encoder to mp4 files. This way you can at least verify your results. At the end you have to implement your own algorithm to encapsulate the stream. But depending on your project (OpenSource?) this might help you.
for a start, look for the commandline tool called mp4box.exe. You can find it on the net. You can use this to convert the files generated with the encoder to mp4 files. This way you can at least verify your results. At the end you have to implement your own algorithm to encapsulate the stream. But depending on your project (OpenSource?) this might help you.