Recherche avancée

Sur d’autres sites (3320)

• GStreamer x264enc and SIGSEGV on Android

  18 juillet 2012, par Cipi

  I am trying to play around with GStreamer-java and it's h.264 encoder (x264enc) on Android.

  Now, I tried to encode raw video from test video source (set its is-live property to TRUE), and when data starts flowing into the x264enc Element, I get SIGSEGV in libx264.so and my app crashes. And it looks like code is dereferencing a NULL... Exact error location is in file "gstreamer_ndk_bundle/x264/encoder/analyse.c" line 963. And it is this line :

  h->mc.memcpy_aligned( h->mb.pic.i8x8_dct_buf, h->dct.luma8x8, sizeof(h->mb.pic.i8x8_dct_buf) );

  My device's CPU is Cortex 8 and it has NEON feature...

  I tried to printout to a file all h structure fields, and no field is uninitialized.

  Android DEBUG info from LogCat : HERE

  My questions :

  1. How can I know what is throwing SIGSEGV, because this line of code seems to be ok ?
  2. Is this a known bug or an expected behavior of x264 encoder on ARM CPU or I am missing some encoder configuration ?
  3. I noticed that if I set property analyse of the encoder to 0x3:0x113 (some random value I found online) I don't get SIGSEGV, encoder seems to work, but I get error in pipeline : Data flow error. Can I just set analyse to a meaningful value ?

  Update

  I fprintf-ed all important "h" fields just before execution runs in x264_mb_analyse_intra() function and just before the line that it breaks, guess it could help : http://pastebin.com/sAfX239j

• GStreamer x264enc and SIGSEGV on Android

  18 juillet 2012, par Cipi

  I am trying to play around with GStreamer-java and it's h.264 encoder (x264enc) on Android.

  Now, I tried to encode raw video from test video source (set its is-live property to TRUE), and when data starts flowing into the x264enc Element, I get SIGSEGV in libx264.so and my app crashes. And it looks like code is dereferencing a NULL... Exact error location is in file "gstreamer_ndk_bundle/x264/encoder/analyse.c" line 963. And it is this line :

  h->mc.memcpy_aligned( h->mb.pic.i8x8_dct_buf, h->dct.luma8x8, sizeof(h->mb.pic.i8x8_dct_buf) );

  My device's CPU is Cortex 8 and it has NEON feature...

  I tried to printout to a file all h structure fields, and no field is uninitialized.

  Android DEBUG info from LogCat : HERE

  My questions :

  1. How can I know what is throwing SIGSEGV, because this line of code seems to be ok ?
  2. Is this a known bug or an expected behavior of x264 encoder on ARM CPU or I am missing some encoder configuration ?
  3. I noticed that if I set property analyse of the encoder to 0x3:0x113 (some random value I found online) I don't get SIGSEGV, encoder seems to work, but I get error in pipeline : Data flow error. Can I just set analyse to a meaningful value ?

  Update

  I fprintf-ed all important "h" fields just before execution runs in x264_mb_analyse_intra() function and just before the line that it breaks, guess it could help : http://pastebin.com/sAfX239j

• FFMPEG - How to Extract Frames As Images While Removing Sequentially Duplicate Frames

  19 mars 2017, par Michael Nelson

  Is there any way (via script or preferably some parameter in calling ffmpeg that I missed) to extract frames from an avi file and ignore sequentially duplicate frames, thus being able to go through the pictures looking only at the deltas/changes ?

  I frequently have to record meetings at work and a lot of the time, the client screen that I am looking at is not changing while we are talking over the phone. After the meeting is over, I need to use these images as part of our documentation and specifications gathering.

  I know that I could just output every frame and run them through any given duplicate file remover utility, but this would remove ALL duplicate frames. So, if the frames extracted went like this :

  A, A, A, B, B, B, B, C, C, A, A, C, C, C, B, B, B ...

  Running them through a typical duplicate file remover, I would get : A, B, C

  What I would want is : A, B, C, A, C, B

  The command that I am currently using to extract the images is :

  ffmpeg.exe -i file.avi -ss 0 -sameq -f image2 -r 1 images%5d.png

  I was getting every frame beforehand (removing the -r 1 from above), but that was generating way too many frames to deal with since these online meetings can go for hours, so for now, I get one frame per second from the file.

  A Windows based solution would be preferable, however, I’m sure other people would be interested in solutions on other platforms if available.

  Any solution or point in the right direction is much appreciated.