I am trying to capture and encode video in H264 to send as RTP (SIP client).

Using the DevVideo API I was able to setup the capture and overall encoding process. However, since the documentation and examples using this API are very few, I am now having problems with the correct way to pack the frames to send as RTP and determining what information the stream and coding standard specific init output have (other specific infos are not generated/accessible).

Ideally I require access to NAL (if generated by the encoder) or at least some way to know the information contained in the stream or other additional info.

I noticed the avc.h header contains a few structures and classes that may contain the information I require but I couldn't find any way to access them, be it either codec or coding standard specific info.

Any help would be appreciated.

Additional info:

Profile/Mime: video/h264; profile-level-id=42800B
Encoder Accelerated: Yes
Encoder Direct Capture: Yes
Input format: Yuv raw data, 176x144
Output: Codec picture, Rtp Payload