PIP Mode for WebRTC video calls in iOS

[suneelreddymunagala]

@mekya, I started working on the PIP functionality for IOS. I finished my rendering with RTCMTLVideoView. Then I realised what I'd done wrong: MTKView/GLKView aren't supported in pip mode. Now I intend to Create frame render to display user's video instead of RTCMTLVideoView/RTCEAGLVideoView . But I can't proceed since I can't convert RTCVideoTrack to RTCVideoFrame. Any code fragments or branches where we can discover the idea/solution would be really appreciated.

/*
1. get videoTrackObject
2. convert RTCVideoTrack to RTCVideoFrame
3. create pixelBuffer using RTCVideoFrame
4. create CMSampleBuffer using CVPixelBuffer
5. update in the localView/remoteview UI using cmSampleBuffer
6. implement PIP related methods.
*/

I referred following links to achieve the pip for video calls:
pip for webrtc video calls in ios
adopting_picture_in_picture_for_video_calls
accessing_the_camera_while_multitasking_on_ipad

[suneelreddymunagala]

@mekya , Any idea on this ?

[mekya]

Hi @suneelreddymunagala ,
Thank you for the reminder.
I don't have the opportunity to take a look at this issue. I've focused on delivering somethings to some customers.
I try to take a look at in my convenient time.

Alternatively, any help from community is greatly appreciated.

Thank you
Oguz