There is a long way between UDP packets and a displayable frame extracted from those packets. The steps are:
- UDP reception
- packet ordering, removal of duplicates, identification of lost packets
- depacketizing into H.264 bitstream (H.264 ES)
- use of H.264 video decoder
- fitting of decoder output to match your needs (texture vs. bytes, conversion to RGB and alike)
The steps are all non-trivial in their own way and in most cases you don't deal with them directly. In most cases there is some assumption and well-known protocol used, then you use existing piece of software to do the job for you taking into account that there is an agreement, such as video streaming per well known specification, and this software is capable to process data respectively.
For example, UDP packets might be structured according to real-time protocol (RTP). The protocol itself is defined by respective A Transport Protocol for Real-Time Applications document, then there is another document RTP Payload Format for H.264 Video for H.264 specifically, then it is assumed that the data is submitted to platform services for H.264 decoding, such as H.264 Video Decoder and so finally there is something approaching to be a video frame in the format you want it.
In all the libraries i've seen which helps to load a stream and render it somehow, they all require a Uri as a source.
Since the libraries (controls, sometimes high level APIs) would typically sole some problem end to end, they are covering streaming needs for all mentioned steps 1-5. You provide a network location (Uri) where the data comes from and can be received, then the library displays the data in user friendly way, displayed video that is.
Your initial description of the problem you are trying to solve
I am constantly receiving bytes (essentially streaming video) for a video that is H.264 encoded...
has no mention of the format of the data. This important and essential part alone makes it impossible to get you any solution for the problem. If you know the details for format and protocol, you mention it and you look for solution exactly for it. If you don't you have to find it and either deal with it yourself such as solving steps 1-3 in your own way and fit the data to more or less standard form that steps 4-5 could be solved in a regular way. If you know the format and there are platform and/or third party solutions for this type of data stream, then you again mention the type of data and you would probably want to use such solutions in your application.