Recherche avancée

Sur d’autres sites (4470)

• Android - Choosing between MediaRecorder, MediaCodec and Ffmpeg

  15 mars 2017, par Rohan Stark

  I am working on a video recording and sharing application for Android. The specifications of the app are as follows :-

  • Recording a 10 second (maximum) video from inside the app (not using the device’s camera app)
  • No further editing on the video
  • Storing the video in a Firebase Cloud Storage (GCS) bucket
  • Downloading and playing of the said video by other users

  From the research, I did on SO and others sources for this, I have found the following (please correct me if I am wrong) :-

  The three options and their respective features are :-

  1.Ffmpeg

  • Capable of achieving the above goal and has extensive answers and explanations on sites like SO, however
  • Increases the APK size by 20-30mb (large library)
  • Runs the risk of not working properly on certain 64-bit devices

  2.MediaRecorder

  • Reliable and supported by most devices
  • Will store files in .mp4 format (unless converted to h264)
  • Easier for playback (no decoding needed)
  • Adds the mp4 and 3gp headers
  • Increases latency according to this question

  3.MediaCodec

  • Low level
  • Will require MediaCodec, MediaMuxer, and MediaExtractor
  • Output in h264 ( without using MediaMuxer for playback )
  • Good for video manipulations (though, not required in my use case)
  • Not supported by pre 4.3 (API 18) devices
  • More difficult to implement and code (my opinion - please correct me if I am wrong)
  • Unavailability of extensive information, tutorials, answers or samples (Bigflake.com being the only exception)

  After spending days on this, I still can’t figure out which approach suits my particular use case. Please elaborate on what I should do for my application. If there’s a completely different approach, then I am open to that as well.

  My biggest criteria are that the video encoding process be as efficient as possible and the video to be stored in the cloud should have the lowest possible space usage without compromising on the video quality.

  Also, I’d be grateful if you could suggest the appropriate format for saving and distributing the video in Firebase Storage, and point me to tutorials or samples of your suggested approach.

  Thank you in advance ! And sorry for the long read.

• Extract frames from a video results in abnormaly more frames for each fps by using ffmpeg

  15 juin 2021, par alanzzz

  I have a job of using ffmpeg to extract the frames averagely from a video, with different fps. I use this command for it.

  


  ffmpeg -i input.mp4 -r specified_fps -q:v 2 image %4d.png

  


  And I have 3 questions about this task.

  


  

  1. What I expect is that if I double the fps, the number of extracted frames will also get doubled. However, that's not the case. Take one of the input videos as an example. I get 2 extra frames for all the sample factors (80 / 158 / 236 / 392 v.s. 78 / 156 / 234 / 390). Does it related to the mechanism of picking/dropping frames when extracting frames from a video？ (credit to @Tom Yan)

  


  


  video info

  


  

  • Duration : 1min18s

  


  • Frame rate mode : constant (CFR)

  


  • Frame rate : 30.0 FPS

  


  • Total number of frames : 2340

  


  


  Config setting & results

  


  





  


  


  


  


  


  



  


  


  


  


  


  


  


  


  


  


  


  


  


  


  


  


  


  


  


  


  FPS	Actual num of frames	correct num of frames
  1	80	78
  2	158	156
  3	236	234
  5	392	390

  


  


  


  2. I check for the output images, the extracted frames for different fps are totally different from each other. In other words, for example, the 1st image for fps=1 is not the same as the 1st image for fps=2. Is that legitimate ? And is it possible for me the get some identical images for different fps ?

     


  


  3. The last problem is that for some videos I use, the difference between the 1st and 2nd image is different from the difference between the 2nd and 3rd. While for the remaining images, the differences become average. To be specific, there is only a slight change from 1st to 2nd frame, while for 2nd to 3rd, 3rd to 4th, and so on, the changes are the same, which is normally distributed according to the specified FPS. I am wondering why such a case happens ? Does it relate to the I-frame, B-frame, P-frame, GOP, or IDR ?

     


  


  


  I am new to this field and cannot find some useful info from other places. I've tried my best to describe my questions clearly. Feel free to leave some comments. Any help would do me a great favor. Thanks in advance !

  


• Is it possible to stream video over RTP without transcoding or compressing input file before transmitting using FFMpeg commandline ?

  11 avril 2017, par Souvik Das

  FFmpeg supports 2 type of RTP payload type : MPEGTS/MP2T (PT 33) & Dynamic (PT 96). Dynamic PT requires explicit SDP presence at receiver while MPEGTS/MP2T doesn’t.
  I used FFmpeg as both transmitter and receiver (with Loopback/localhost) and compared PSNR of the respective streams :

  Case 1 : FFmpeg Dynamic RTP

  Sender:
  
  
  
      ffmpeg -re -i 'sample.avi' -c:a copy -c:v copy -f rtp -y 'rtp://@225.0.0.1:5555' > sample.sdp
  
  
  
  Receiver:
  
  
  
      ffmpeg -protocol_whitelist file,udp,rtcp,rtp -i sample.sdp -y rec.ts
  
  
  
  Result:
  
  
  
      PSNR avg. = 38
  
  
  
  This means that in idle condition, we are still not getting a perfect stream. I suspect, it's because Transcoding still takes place which downgrades the quality of video before transmitting at sender.

  Case 2 : FFmpeg MPEGTS RTP

  Sender:
  
  
  
      ffmpeg -re -i 'sample.avi' -c:a copy -c:v copy -f rtp_mpegts -y 'rtp://@225.0.0.1:5555'
  
  
  
  Receiver:
  
  
  
  
  
      ffmpeg -protocol_whitelist file,udp,rtcp,rtp -i sample.sdp -f mpegts -y rec.ts
  
  
  
  Result:
  
  
  
      Large # Frame Losses!
  
  
  
  So, at Receiver, I used VLC for recording the streams. Although there was no/negligible frame loss, but PSNR avg. = 18 !!

  Earlier in a dedicated VLC Streamer & Recorder test, when the same video was streamed, PSNR avg. = Infinity (No Quality Loss). I want to shift to FFMpeg alternative for streaming because, I want to introduce some programmability factors for a sophisticated research work.

  Hence, It would be really great, if somebody could provide me some input as to how I can achieve uncompressed & lossless video streaming using FFMpeg over RTP.

  Notes :

  1. I must use RTP only. I can't use RTSP or other streaming methods incl. direct UDP (udp://)
  
  2. VLC Media Player / Libvlc used in this case, also used RTP for all cases.
  
  3. It can assumed that Streamer and Recorder are present on same disk or has same access to storage.
  
  4. Must support multicast!