I'm using the ffmpeg library in an application to receive streams from
cameras via RTSP. However, since these cameras are on a network with poor
bandwidth (a weak wi-fi connection), I'm having issues where the video feed
appears to 'smear' along the bottom of the frame.
I'm confident that the cause of this 'smearing' is RTP packets being
the network and leading to corrupted IDR frames. Unfortunately, using tcp
transport with RTSP is not an option as the video is somewhat latency
Looking over the help pages for my local ffmpeg binary (ffmpeg version
3.4.6-0ubuntu0.18.04.1), I noticed that the h264 decoder has an "enable_er"
option that supposedly enables some form of 'error resilience' in the
I have two questions:
1. What does the "enable_er" option of the h264 decoder actually do?
2. Are there any other options available that might help mask corrupted IDR