Recherche avancée

Sur d’autres sites (4902)

• Is there an efficient way to retrieve frames from a video in Android ?

  28 mars 2015, par Naveed

  I have an app which requires me to retrieve frames from a video and do some processing with them. However it seems like that the frame retrieval is very slow to the point where it is unacceptable. Sometimes it is taking upto 2.5 second to retrieve a single frame. I am using the MediaMetadataRetriever as most stackoverflow questions suggested. However the performance is very bad. Here is what I have :

     private List<bitmap> retrieveFrames() {
  
  
  
          MediaMetadataRetriever fmmr = new MediaMetadataRetriever();
  
          fmmr.setDataSource("/path/to/some/video.mp4");
  
          String strLength = fmmr.extractMetadata(MediaMetadataRetriever.METADATA_KEY_DURATION);
  
          long milliSecs = Long.parseLong(strLength);
  
          long microSecLength = milliSecs * 1000;
  
  
  
          Log.d("TAG", "length: " + microSecLength);
  
          long one_sec = 1000000; // one sec in micro seconds
  
  
  
          ArrayList<bitmap> frames = new ArrayList&lt;>();
  
          int j = 0;
  
          for (int i = 0; i &lt; microSecLength; i += (one_sec / 5)) {
  
              long time = System.currentTimeMillis();
  
              Bitmap frame = fmmr.getFrameAtTime(i, MediaMetadataRetriever.OPTION_CLOSEST);
  
              j++;
  
              Log.d("TAG", "Frame number: " + j + " Time taken: " + (System.currentTimeMillis() - time));
  
              // commented out because each frame would be written to disk instead of holding them in memory
  
              //  frames.add(frame);
  
          }
  
          fmmr.release();
  
          return frames;
  
      }
  
  </bitmap></bitmap>

  The above will logs :

  03-26 21:49:29.781  13213-13239/com.example.naveed.myapplication D/TAG﹕ length: 4949000
  
  03-26 21:49:30.187  13213-13239/com.example.naveed.myapplication D/TAG﹕ Frame number: 1 Time taken: 406
  
  03-26 21:49:30.779  13213-13239/com.example.naveed.myapplication D/TAG﹕ Frame number: 2 Time taken: 592
  
  03-26 21:49:31.578  13213-13239/com.example.naveed.myapplication D/TAG﹕ Frame number: 3 Time taken: 799
  
  03-26 21:49:32.632  13213-13239/com.example.naveed.myapplication D/TAG﹕ Frame number: 4 Time taken: 1054
  
  03-26 21:49:33.895  13213-13239/com.example.naveed.myapplication D/TAG﹕ Frame number: 5 Time taken: 1262
  
  03-26 21:49:35.382  13213-13239/com.example.naveed.myapplication D/TAG﹕ Frame number: 6 Time taken: 1486
  
  03-26 21:49:37.128  13213-13239/com.example.naveed.myapplication D/TAG﹕ Frame number: 7 Time taken: 1746
  
  03-26 21:49:39.077  13213-13239/com.example.naveed.myapplication D/TAG﹕ Frame number: 8 Time taken: 1948
  
  03-26 21:49:41.287  13213-13239/com.example.naveed.myapplication D/TAG﹕ Frame number: 9 Time taken: 2210
  
  03-26 21:49:43.717  13213-13239/com.example.naveed.myapplication D/TAG﹕ Frame number: 10 Time taken: 2429
  
  03-26 21:49:44.093  13213-13239/com.example.naveed.myapplication D/TAG﹕ Frame number: 11 Time taken: 376
  
  03-26 21:49:44.707  13213-13239/com.example.naveed.myapplication D/TAG﹕ Frame number: 12 Time taken: 614
  
  03-26 21:49:45.539  13213-13239/com.example.naveed.myapplication D/TAG﹕ Frame number: 13 Time taken: 831
  
  03-26 21:49:46.597  13213-13239/com.example.naveed.myapplication D/TAG﹕ Frame number: 14 Time taken: 1057
  
  03-26 21:49:47.875  13213-13239/com.example.naveed.myapplication D/TAG﹕ Frame number: 15 Time taken: 1278
  
  03-26 21:49:49.384  13213-13239/com.example.naveed.myapplication D/TAG﹕ Frame number: 16 Time taken: 1508
  
  03-26 21:49:51.112  13213-13239/com.example.naveed.myapplication D/TAG﹕ Frame number: 17 Time taken: 1728
  
  03-26 21:49:53.096  13213-13239/com.example.naveed.myapplication D/TAG﹕ Frame number: 18 Time taken: 1983
  
  03-26 21:49:55.315  13213-13239/com.example.naveed.myapplication D/TAG﹕ Frame number: 19 Time taken: 2218
  
  03-26 21:49:57.711  13213-13239/com.example.naveed.myapplication D/TAG﹕ Frame number: 20 Time taken: 2396
  
  03-26 21:49:58.065  13213-13239/com.example.naveed.myapplication D/TAG﹕ Frame number: 21 Time taken: 354
  
  03-26 21:49:58.640  13213-13239/com.example.naveed.myapplication D/TAG﹕ Frame number: 22 Time taken: 574
  
  03-26 21:49:59.369  13213-13239/com.example.naveed.myapplication D/TAG﹕ Frame number: 23 Time taken: 728
  
  03-26 21:50:00.112  13213-13239/com.example.naveed.myapplication D/TAG﹕ Frame number: 24 Time taken: 742
  
  03-26 21:50:00.834  13213-13239/com.example.naveed.myapplication D/TAG﹕ Frame number: 25 Time taken: 721

  As you can see from above, it is taking about 18 - 25 sec to retrieve 25 frames from a 4 sec long video.

  I have also tried this which uses FFmpeg underneath to do the same. I am not sure how well this library is implemented but it only improves the over all performance by a couple of seconds meaning it takes about 15-20 sec to do the same.

  So my question is : is there a way to do it quicker ? My friend has an iOS app where he does something similar but it only takes couple of seconds and he is grabbing even more frames however he is not sure how to do it on android.

  Is there anything on android that would speed up the process. Am I approaching this wrong ?

  The end goal is to stitch those frames together into a gif.

• Why a batch processing of ffmpeg is freezing the system ?

  3 septembre 2019, par Krishna Chebrolu

  I have a requirement of splitting smaller chunks of videos from 50+ mp4 source files for 5000+ records. Each record may result in 2 or 3 smaller chunks from as many source files out of those 50+.

  The logic to determine which source file to be picked up is written in Java and then fed to ffmpeg on Runtime.getRuntime().exec() using ExecutorService with newFixedThreadPool as below :

  private static boolean processqueue(ArrayList<string> cmds) {
  
      final ExecutorService pool;
  
      int threadsnum = Runtime.getRuntime().availableProcessors()-2;
  
      pool = Executors.newFixedThreadPool(threadsnum);
  
  
  
      for(final String cmd: cmds){ 
  
          pool.execute(new Runnable() {
  
              public void run() {
  
                  System.out.println(cmd);
  
                  try {
  
                      Runtime.getRuntime().exec(cmd);
  
                  } catch (IOException e) {
  
                      e.printStackTrace();
  
                      pool.shutdown();
  
                  }
  
              }
  
          });
  
      }                   
  
      pool.shutdown();
  
  
  
      // wait for them to finish for up to one minute.
  
      try {
  
          if(!pool.awaitTermination(1, TimeUnit.MINUTES)) {
  
              pool.shutdownNow();
  
          }
  
  
  
          //Wait a while for tasks to respond to being cancelled
  
          if(!pool.awaitTermination(1, TimeUnit.MINUTES))
  
              System.err.println("Pool did not shutdown properly");
  
  
  
      } catch (InterruptedException e) {
  
          e.printStackTrace();
  
          pool.shutdownNow();
  
          //Preserve interrupt status
  
          Thread.currentThread().interrupt();
  
          return false;
  
      }                   
  
  
  
      return true;
  
  }
  
  </string>

  the String cmd value is one of these based on split or merge requirement :

  for split :

  ffmpeg -y -ss 00:00:00 -t 00:08 -i E:/tmp/fin12.mp4 -acodec copy -vcodec copy E:/tmp/Intermed/0136f.mp4

  or

  for merge :

  ffmpeg -y -i E:/tmp/Inter/0136c0.mp4 -i E:/tmp/Inter/0136c1.mp4 -i E:/tmp/Inter/0136f.mp4 -i E:/tmp/Jingle.mp4 -i E:/tmp/wm1280.png -filter_complex "[0:v][0:a][1:v][1:a][2:v][2:a][3:v][3:a]concat=n=4:v=1:a=1[vv][a];[vv][4:v]overlay=x=0:y=H-overlay_h[v]" -map "[v]" -map "[a]" E:/tmp/final/0136.mp4

  On first attempt, only 250 records were processed. And, on subsequent attempt of balance records processing, it threw below exception ; but, processed another 300 records :

  java.io.IOException: Cannot run program "ffmpeg": CreateProcess error=1455, The paging file is too small for this operation to complete
  
  at java.lang.ProcessBuilder.start(Unknown Source)

  And, this code freezes often. Why is ExecutorService not holding up the queue to process all the records and exit gracefully ? What am I doing wrong ?

  Note : I’m calling Java class from windows batch script by passing relevant arguments which is executed from command line.

• Dreamcast Track Sizes

  1er mars 2015, par Multimedia Mike — Sega Dreamcast

  I’ve been playing around with Sega Dreamcast discs lately. Not playing the games on the DC discs, of course, just studying their structure. To review, the Sega Dreamcast game console used special optical discs named GD-ROMs, where the GD stands for “gigadisc”. They are capable of holding about 1 gigabyte of data.

  You know what’s weird about these discs ? Each one manages to actually store a gigabyte of data. Each disc has a CD portion and a GD portion. The CD portion occupies the first 45000 sectors and can be read in any standard CD drive. This area is divided between a brief data track and a brief (usually) audio track.

  The GD region starts at sector 45000. Sometimes, it’s just one humongous data track that consumes the entire GD region. More often, however, the data track is split between the first track and the last track in the region and there are 1 or more audio tracks in between. But the weird thing is, the GD region is always full. I made a study of it (click for a larger, interactive graph) :

  
  
  

  Some discs put special data or audio bonuses in the CD region for players to discover. But every disc manages to fill out the GD region. I checked up on a lot of those audio tracks that divide the GD data and they’re legitimate music tracks. So what’s the motivation ? Why would the data track be split in 2 pieces like that ?

  I eventually realized that I probably answered this question in this blog post from 4 years ago. The read speed from the outside of an optical disc is higher than the inside of the same disc. When I inspect the outer data tracks of some of these discs, sure enough, there seem to be timing-sensitive multimedia FMV files living on the outer stretches.

  One day, I’ll write a utility to take apart the split ISO-9660 filesystem offset from a weird sector.