[libcamera-devel] Porting libCamera in to RDKC

[Dubey, Vijayanand]

Hi,
This is Vijay form RDK for Camera (RDKC) team in Comcast.
We are trying to understand how can we fit libCamera in our current architecture. In our current RDKC architecture as shown in the block diagram,

[cid:image001.png at 01D63B4B.C3E9BE30]

camera  has process called xStreamer. It runs as a web server (producer )and provides H264 /PCM/G711/AAC/YUV data to client applications (e.g.;  live, cvr, video analytics etc.). Moreover it

  *   Controls the encoding parameter
  *   control single and multi-stream encoding
  *   Overlay on stream
Camera has separate processes  for Live, CVR and video analytics  which connect to xStreamer and send the request to configure the encoding  parameter and get the encoded or RAW AV data over socket.
Based on our initial investigation of libCamera  architecture I have couple of questions
It seems current architecture of  libCamera requires to have a single process with  all functionalities(Live/CVR/Analytics etc.) Is our understanding correct? If yes is their plan to support the RDKC use case in libCamera? In no how can we align RDKC architecture with libCamera?

Regards
Vijay

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.libcamera.org/pipermail/libcamera-devel/attachments/20200605/98794042/attachment-0001.htm>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: image001.png
Type: image/png
Size: 616054 bytes
Desc: image001.png
URL: <https://lists.libcamera.org/pipermail/libcamera-devel/attachments/20200605/98794042/attachment-0001.png>