I am trying to parse various mpeg4 frames from an rtp stream coming from an axial camera and transfer packets to the ffmpeg library using the avcodec_decode_video function. here are the steps that I am doing 1. the rtsp stream is initialized 2. the rtp stream starts to flow in 3. The first packet that I receive starts with 000001b0 ... and the configuration data follows, and after that frame starts with 000001b6 .. the second payload rtp will be different until I get the rtp packet where the marker bit is set. after that I get a packet starting with 000001b6 and goes for about 5-10 RTT packets. This pattern is repeated.
what do I do if I detect 000001b0 / b6 - I will accumulate all packets coming after and feed a larger buffer to the avcodec_decode_video libavcodec function, after the decoder context is correctly initialized.
But I get a crappy image here, with the uppermost part, a horizontal strip - a crystal clear image, and the rest is crappy. Iām not sure why he behaves like this. Please help me
The data that I receive in the rtp package is dynamic-96.
indicate that: when I transfer iframes and p frames that are absorbed in the propositional protocol of some other manufacturer, ffmpeg can parse and give very good things.
Any help is appreciated.
source
share