Recherche avancée

Sur d’autres sites (11433)

• avutil/half2float : move tables to header-internal structs

  9 août 2022, par Timo Rothenpieler

  avutil/half2float : move tables to header-internal structs
  Having to put the knowledge of the size of those arrays into a multitude
  
  of places is rather smelly.
  

  • [DH] libavcodec/exr.c
  • [DH] libavcodec/exrenc.c
  • [DH] libavcodec/pnm.h
  • [DH] libavcodec/pnmdec.c
  • [DH] libavcodec/pnmenc.c
  • [DH] libavutil/float2half.h
  • [DH] libavutil/half2float.h

• ffmpeg replacing audio track with track from another video but different length (sync audio)

  27 août 2019, par Matt

  Background

  I have digitized some old Canon Video8 tapes. I used a SONY Digital8 camera (which is backwards compatible with Video8) to output DV (using it’s inbuilt ADC). The conversion process worked well for the video but the audio came through jumpy/distorted in places. This left me with a problem, was it the Camera or the tape ? So I bought another Samsung Video8 camera (just analog) and using the SONY’s passthrough feature output from the Samsung (Composite & mono audio) into the SONY which output the DV. Much to my delight it worked ! The audio was clear.

  Result

  DV01.avi - Good Video / Crap Audio
  
  DV02.avi - Crap Video / Good Audio

  Ok so obviously what I would like to do is take the video track from DV01 and the audio track from DV02 and join/mux ? them WITHOUT re-encoding.

  Problem 1 : They have different start times so just copying over the audio track will result it not being in sync.

  After some googling I found you can use ffmpeg to take care of this :

  Firstly here is the Video info using : ffmpeg -i DV01.avi

  Input #0, avi, from 'DV01.avi':
  
    Duration: 02:53:06.68, start: 0.000000, bitrate: 28878 kb/s
  
      Stream #0:0: Video: dvvideo, yuv420p, 720x576 [SAR 16:15 DAR 4:3], 25000 kb/s, 25 fps, 25 tbr, 25 tbn, 25 tbc
  
      Stream #0:1: Audio: pcm_s16le, 32000 Hz, stereo, s16, 1024 kb/s
  
      Stream #0:2: Audio: pcm_s16le, 32000 Hz, stereo, s16, 1024 kb/s

  Muxing :

  ffmpeg -itsoffset 4 -i DV01.avi -i DV02.avi -map 0:v -map 1:a -c copy output.avi

  In the example above I am delaying the start of the video 4 seconds (I think ?) This is just an example I haven’t actually tried it as the file sizes are 40Gb !

  So my QUESTION is :

  Given the background above what would be the best way to join/mux the two streams together (without re-encoding) with the audio being in sync. Given that syncing the audio to the video may need millisecond tweaking I don’t believe trial and error is a good idea (I don’t want to tweak it by 10ms then rinse and repeat for a 40Gb file) ?

  I just had a thought, could I say create a 10 second clip (from the start) of each video and use them to find the reference/sync start point then use that when muxing the 40Gb versions ?

  Anyway, you get the idea. Looking for ways to solve this problem. Thanks !

• aarch64 : vp9itxfm : Do separate functions for half/quarter idct16 and idct32

  22 novembre 2016, par Martin Storsjö

  aarch64 : vp9itxfm : Do separate functions for half/quarter idct16 and idct32
  This work is sponsored by, and copyright, Google.
  This avoids loading and calculating coefficients that we know will
  
  be zero, and avoids filling the temp buffer with zeros in places
  
  where we know the second pass won’t read.
  This gives a pretty substantial speedup for the smaller subpartitions.
  The code size increases from 14740 bytes to 24292 bytes.
  The idct16/32_end macros are moved above the individual functions ; the
  
  instructions themselves are unchanged, but since new functions are added
  
  at the same place where the code is moved from, the diff looks rather
  
  messy.
  Before :
  
  vp9_inv_dct_dct_16x16_sub1_add_neon :     236.7
  
  vp9_inv_dct_dct_16x16_sub2_add_neon :    1051.0
  
  vp9_inv_dct_dct_16x16_sub4_add_neon :    1051.0
  
  vp9_inv_dct_dct_16x16_sub8_add_neon :    1051.0
  
  vp9_inv_dct_dct_16x16_sub12_add_neon :   1387.4
  
  vp9_inv_dct_dct_16x16_sub16_add_neon :   1387.6
  
  vp9_inv_dct_dct_32x32_sub1_add_neon :     554.1
  
  vp9_inv_dct_dct_32x32_sub2_add_neon :    5198.5
  
  vp9_inv_dct_dct_32x32_sub4_add_neon :    5198.6
  
  vp9_inv_dct_dct_32x32_sub8_add_neon :    5196.3
  
  vp9_inv_dct_dct_32x32_sub12_add_neon :   6183.4
  
  vp9_inv_dct_dct_32x32_sub16_add_neon :   6174.3
  
  vp9_inv_dct_dct_32x32_sub20_add_neon :   7151.4
  
  vp9_inv_dct_dct_32x32_sub24_add_neon :   7145.3
  
  vp9_inv_dct_dct_32x32_sub28_add_neon :   8119.3
  
  vp9_inv_dct_dct_32x32_sub32_add_neon :   8118.7
  After :
  
  vp9_inv_dct_dct_16x16_sub1_add_neon :     236.7
  
  vp9_inv_dct_dct_16x16_sub2_add_neon :     640.8
  
  vp9_inv_dct_dct_16x16_sub4_add_neon :     639.0
  
  vp9_inv_dct_dct_16x16_sub8_add_neon :     842.0
  
  vp9_inv_dct_dct_16x16_sub12_add_neon :   1388.3
  
  vp9_inv_dct_dct_16x16_sub16_add_neon :   1389.3
  
  vp9_inv_dct_dct_32x32_sub1_add_neon :     554.1
  
  vp9_inv_dct_dct_32x32_sub2_add_neon :    3685.5
  
  vp9_inv_dct_dct_32x32_sub4_add_neon :    3685.1
  
  vp9_inv_dct_dct_32x32_sub8_add_neon :    3684.4
  
  vp9_inv_dct_dct_32x32_sub12_add_neon :   5312.2
  
  vp9_inv_dct_dct_32x32_sub16_add_neon :   5315.4
  
  vp9_inv_dct_dct_32x32_sub20_add_neon :   7154.9
  
  vp9_inv_dct_dct_32x32_sub24_add_neon :   7154.5
  
  vp9_inv_dct_dct_32x32_sub28_add_neon :   8126.6
  
  vp9_inv_dct_dct_32x32_sub32_add_neon :   8127.2
  Signed-off-by : Martin Storsjö <martin@martin.st>
  

  • [DBH] libavcodec/aarch64/vp9itxfm_neon.S