Sur d’autres sites (8096)

• Android : Recording and Streaming at the same time

  23 avril 2020, par Bruno Siqueira

  This is not really a question as much as it is a presentation of all my attempts to solve one of the most challenging functionalities I was faced with.

  



  I use libstreaming library to stream realtime videos to Wowza Server and I need to record it at the same time inside the SD card. I am presenting below all my attempts in order to collect new ideias from the community.

  



  Copy bytes from libstreaming stream to a mp4 file

  



  Development

  



  We created an interception in libstreaming library to copy all the sent bytes to a mp4 file. Libstreaming sends the bytes to Wowza server through a LocalSocket. It users MediaRecorder to access the camera and the mic of the device and sets the output file as the LocalSocket's input stream. What we do is create a wrapper around this input stream extending from InputStream and create a File output stream inside it. So, every time libstreaming executes a reading over the LocaSocket's input stream, we copy all the data to the output stream, trying to create a valid MP4 file.

  



  Impediment

  



  When we tried to read the file, it is corrupted. We realized that there are meta information missing from the MP4 file. Specifically the moov atom. We tried to delay the closing of the streaming in order to give time to send this header (this was still a guessing) but it didn't work. To test the coherence of this data, we used a paid software to try to recover the video, including the header. It became playable, but it was mostly green screen. So this became an not trustable solution. We also tried using "untrunc", a free open source command line program and it couldn't even start the recovery, since there was no moov atom.

  



  Use ffmpeg compiled to android to access the camera

  



  Development

  



  FFMPEG has a gradle plugin with a java interface to use it inside Android apps. We thought we could access the camera via command line (it is probably in "/dev/video0") and sent it to the media server.

  



  Impediment

  



  We got the error "Permission Denied" when trying to access the camera. The workaround would be to root the device to have access to it, but it make the phones loose their warranty and could brick them.

  



  Use ffmpeg compiled to android combined with MediaRecorder

  



  Development

  



  We tried to make FFMPEG stream a mp4 file being recorded inside the phone via MediaRecorder

  



  Impediment

  



  FFMPEG can not stream MP4 files that are not yet done with the recording.

  



  Use ffmpeg compiled to android with libstreaming

  



  Development

  



  Libstreaming uses LocalServerSocket as the connection between the app and the server, so we thought that we could use ffmpeg connected with LocalServerSocket local address to copy the streaming directly to a local file inside the SD card. Right after the streaming started, we also ran the ffmpeg command to start recording the data to a file. Using ffmpeg, we believed that it would create a MP4 file in the proper way, which means with the moov atom header included.

  



  Impediment

  



  The "address" created is not readable via command line, as a local address inside the phone. So the copy is not possible.

  



  Use OpenCV

  



  Development

  



  OpenCV is an open-source, cross-platform library that provides building blocks for computer vision experiments and applications. It offers high-level interfaces for capturing, processing, and presenting image data. It has their own APIs to connect with the device camera so we started studding it to see if it had the necessary functionalities to stream and record at the same time.

  



  Impediment

  



  We found out that the library is not really defined to do this, but more as image mathematical manipulation. We got even the recommendation to use libstreaming (which we do already).

  



  Use Kickflip SDK

  



  Development

  



  Kickflip is a media streaming service that provides their own SDK for development in android and IOS. It also uses HLS instead of RTMP, which is a newer protocol.

  



  Impediment

  



  Their SDK requires that we create a Activity with camera view that occupies the entire screen of the device, breaking the usability of our app.

  



  Use Adobe Air

  



  Development

  



  We started consulting other developers of app's already available in the Play Store, that stream to servers already.

  



  Impediment

  



  Getting in touch with those developers, they reassured that would not be possible to record and stream at the same time using this technology. What's more, we would have to redo the entire app from scratch using Adobe Air.

  



  UPDATE

  



  Webrtc

  



  Development

  



  We started using WebRTC following this great project. We included the signaling server in our NODEJS server and started doing the standard handshake via socket. We were still toggling between local recording and streaming via webrtc.

  



  Impediment

  



  Webrtc does not work in every network configuration. Other than that, the camera acquirement is all native code, which makes a lot harder to try to copy the bytes or intercept it.

  


• Android : Recording and Streaming at the same time

  29 mars 2016, par Bruno Siqueira

  This is not really a question as much as it is a presentation of all my attempts to solve one of the most challenging functionalities I was faced with.

  I use libstreaming library to stream realtime videos to Wowza Server and I need to record it at the same time inside the SD card. I am presenting below all my attempts in order to collect new ideias from the community.

  Copy bytes from libstreaming stream to a mp4 file

  Development

  We created an interception in libstreaming library to copy all the sent bytes to a mp4 file. Libstreaming sends the bytes to Wowza server through a LocalSocket. It users MediaRecorder to access the camera and the mic of the device and sets the output file as the LocalSocket’s input stream. What we do is create a wrapper around this input stream extending from InputStream and create a File output stream inside it. So, every time libstreaming executes a reading over the LocaSocket’s input stream, we copy all the data to the output stream, trying to create a valid MP4 file.

  Impediment

  When we tried to read the file, it is corrupted. We realized that there are meta information missing from the MP4 file. Specifically the moov atom. We tried to delay the closing of the streaming in order to give time to send this header (this was still a guessing) but it didn’t work. To test the coherence of this data, we used a paid software to try to recover the video, including the header. It became playable, but it was mostly green screen. So this became an not trustable solution. We also tried using "untrunc", a free open source command line program and it couldn’t even start the recovery, since there was no moov atom.

  Use ffmpeg compiled to android to access the camera

  Development

  FFMPEG has a gradle plugin with a java interface to use it inside Android apps. We thought we could access the camera via command line (it is probably in "/dev/video0") and sent it to the media server.

  Impediment

  We got the error "Permission Denied" when trying to access the camera. The workaround would be to root the device to have access to it, but it make the phones loose their warranty and could brick them.

  Use ffmpeg compiled to android combined with MediaRecorder

  Development

  We tried to make FFMPEG stream a mp4 file being recorded inside the phone via MediaRecorder

  Impediment

  FFMPEG can not stream MP4 files that are not yet done with the recording.

  Use ffmpeg compiled to android with libstreaming

  Development

  Libstreaming uses LocalServerSocket as the connection between the app and the server, so we thought that we could use ffmpeg connected with LocalServerSocket local address to copy the streaming directly to a local file inside the SD card. Right after the streaming started, we also ran the ffmpeg command to start recording the data to a file. Using ffmpeg, we believed that it would create a MP4 file in the proper way, which means with the moov atom header included.

  Impediment

  The "address" created is not readable via command line, as a local address inside the phone. So the copy is not possible.

  Use OpenCV

  Development

  OpenCV is an open-source, cross-platform library that provides building blocks for computer vision experiments and applications. It offers high-level interfaces for capturing, processing, and presenting image data. It has their own APIs to connect with the device camera so we started studding it to see if it had the necessary functionalities to stream and record at the same time.

  Impediment

  We found out that the library is not really defined to do this, but more as image mathematical manipulation. We got even the recommendation to use libstreaming (which we do already).

  Use Kickflip SDK

  Development

  Kickflip is a media streaming service that provides their own SDK for development in android and IOS. It also uses HLS instead of RTMP, which is a newer protocol.

  Impediment

  Their SDK requires that we create a Activity with camera view that occupies the entire screen of the device, breaking the usability of our app.

  Use Adobe Air

  Development

  We started consulting other developers of app’s already available in the Play Store, that stream to servers already.

  Impediment

  Getting in touch with those developers, they reassured that would not be possible to record and stream at the same time using this technology. What’s more, we would have to redo the entire app from scratch using Adobe Air.

  UPDATE

  Webrtc

  Development

  We started using WebRTC following this great project. We included the signaling server in our NODEJS server and started doing the standard handshake via socket. We were still toggling between local recording and streaming via webrtc.

  Impediment

  Webrtc does not work in every network configuration. Other than that, the camera acquirement is all native code, which makes a lot harder to try to copy the bytes or intercept it.

• VP8 Codec SDK "Aylesbury" Release

  28 octobre 2010, par noreply@blogger.com (John Luther)

  Today we’re making available "Aylesbury," our first named release of libvpx, the VP8 codec SDK. VP8 is the video codec used in WebM. Note that the VP8 specification has not changed, only the SDK.

  What’s an Aylesbury ? It’s a breed of duck. We like ducks, so we plan to use duck-related names for each major libvpx release, in alphabetical order. Our goal is to have one named release of libvpx per calendar quarter, each with a theme.

  You can download the Aylesbury libvpx release from our Downloads page or check it out of our Git repository and build it yourself. In the coming days Aylesbury will be integrated into all of the WebM project components (DirectShow filters, QuickTime plugins, etc.). We encourage anyone using our components to upgrade to the Aylesbury releases.

  For Aylesbury the theme was faster decoder, better encoder. We used our May 19, 2010 launch release of libvpx as the benchmark. We’re very happy with the results (see graphs below) :
  

  • 20-40% (average 28%) improvement in libvpx decoder speed
  • Over 7% overall PSNR improvement (6.3% SSIM) in VP8 "best" quality encoding mode, and up to 60% improvement on very noisy, still or slow moving source video.

  

  
  

  

  
  

  

  
  

  The main improvements to the decoder are :
  

  • Single-core assembly "hot spot" optimizations, including improved vp8_sixtap_predict() and SSE2 loopfilter functions
  • Threading improvements for more efficient use of multiple processor cores
  • Improved memory handling and reduced footprint
  • Combining IDCT and reconstruction steps
  • SSSE3 usage in functions where appropriate

  On the encoder front, we concentrated on clips in the 30-45 dB range and saw the biggest gains in higher-quality source clips (greater that 38 dB), low to medium-motion clips, and clips with noisy source material. Many code contributions made this possible, but a few of the highlights were :
  

  • Adaptive width and strength alternate reference frame noise suppression filter with optional motion compensation.
  • Transform improvements (improved accuracy and reduction in round trip error)
  • Trellis-based quantized coefficient optimization
  • Two-pass rate control and quantizer changes
  • Rate distortion changes
  • Zero bin and rounding changes
  • Work on MB-level quality control and bit allocation

  We’re targeting Q1 2011 for the next named libvpx release, which we’re calling Bali. The theme for that release will be faster encoder. We are constantly working on improvements to video quality in the encoder, so after Aylesbury we won’t tie that work to specific named releases.

  WebM at Streaming Media West

  Members of the WebM project will discuss Aylesbury during a session at the Streaming Media West conference on November 3rd (session C203 : WebM Open Video Project Update). For more information, visit www.streamingmedia.com/west.

  John Luther is Product Manager of the WebM Project.