Recherche avancée

Sur d’autres sites (8313)

• Saving Raw Uncompressed Video Files using OpenCv, Gstreamer, and/or FFMPEG ?

  20 septembre 2022, par adav0033

  I have been trying to implement the cv::VideoWriter function from OpenCV to generate a an uncompressed (raw) video file. I started this because of a statement within the OpenCV Documentation which I will link here along with the statement.

  


  cv::VideoWriter::VideoWriter    (   const String &  filename,
int     fourcc,
double  fps,
Size    frameSize,
bool    isColor = true 
)       


  


  "If FFMPEG is enabled, using codec=0 ; fps=0 ; you can create an uncompressed (raw) video file."

  


  Ref. https://docs.opencv.org/3.4/dd/d9e/classcv_1_1VideoWriter.html

  


  However whilst troubleshooting the function I came across the refuting statement,

  


  " VideoCapture and VideoWriter do not provide interface to access raw compressed video stream, except maybe MJPEG in some cases.
Make sure you actually use FFmpeg backend by setting apiPreference parameter : VideoWriter("outfile.avi", cv2.CAP_FFMPEG, ...)"

  


  Ref. https://github.com/opencv/opencv/issues/14573

  


  I am now confused about how I go about writing the cv::VideoWriter function to satisfy the requirements to create a raw uncompressed video file (.avi) and if it is even possible. If it is not possible how do I achieve the outcome of saving an raw uncompressed video file, as I assume it would use some combination of FFMPEG, OpenCV,or Gstreamer.

  


  Note : My code is implemented in c++

  


• ffmpeg - how can I get a specific channel using channel masks (e.g. : get only surround left channel)

  26 novembre 2013, par awfulcode

  I'm trying to build a mixer that allows users to toggle specific channels on and off. I know that different codecs encode those channels in different orders (it's not as simple as getting channel #3 as the center channel every time), so the challenge is to properly identify the channel position in surround audio tracks.

  ffmpeg does list the in the beginning of libavutil/channel_layout.h :

  #define AV_CH_FRONT_LEFT             0x00000001
  
  #define AV_CH_FRONT_RIGHT            0x00000002
  
  #define AV_CH_FRONT_CENTER           0x00000004
  
  #define AV_CH_LOW_FREQUENCY          0x00000008
  
  #define AV_CH_BACK_LEFT              0x00000010
  
  #define AV_CH_BACK_RIGHT             0x00000020
  
  #define AV_CH_FRONT_LEFT_OF_CENTER   0x00000040
  
  #define AV_CH_FRONT_RIGHT_OF_CENTER  0x00000080
  
  #define AV_CH_BACK_CENTER            0x00000100
  
  #define AV_CH_SIDE_LEFT              0x00000200
  
  #define AV_CH_SIDE_RIGHT             0x00000400
  
  #define AV_CH_TOP_CENTER             0x00000800
  
  #define AV_CH_TOP_FRONT_LEFT         0x00001000
  
  #define AV_CH_TOP_FRONT_CENTER       0x00002000
  
  #define AV_CH_TOP_FRONT_RIGHT        0x00004000
  
  #define AV_CH_TOP_BACK_LEFT          0x00008000
  
  #define AV_CH_TOP_BACK_CENTER        0x00010000
  
  #define AV_CH_TOP_BACK_RIGHT         0x00020000
  
  #define AV_CH_STEREO_LEFT            0x20000000  ///< Stereo downmix.
  
  #define AV_CH_STEREO_RIGHT           0x40000000  ///< See AV_CH_STEREO_LEFT.
  
  #define AV_CH_WIDE_LEFT              0x0000000080000000ULL
  
  #define AV_CH_WIDE_RIGHT             0x0000000100000000ULL
  
  #define AV_CH_SURROUND_DIRECT_LEFT   0x0000000200000000ULL
  
  #define AV_CH_SURROUND_DIRECT_RIGHT  0x0000000400000000ULL
  
  #define AV_CH_LOW_FREQUENCY_2        0x0000000800000000ULL

  But looking over the header files and whatever little documentation about ffmpeg I couldn't find anything that tells me how to assign a specific channel position (or mask, as it's described in the header file) to a specific bus in a mixer audio unit.

  Do you know how to do that ? Please let me know if you want to look at the code I'm using right now.

  And as always, thank you so much.

• configure : speed up flatten_extralibs_wrapper()

  30 juillet 2018, par Avi Halachmi (:avih)

  configure : speed up flatten_extralibs_wrapper()
  x50 - x200 faster.
  Currently configure spends 50-70% of its runtime inside a single
  
  function : flatten_extralibs[_wrapper] - which does string processing.
  During its run, nearly 20K command substitutions (subshells) are used,
  
  including its callees unique() and resolve(), which is the reason
  
  for its lengthy run.
  This commit avoids all subshells during its execution, speeding it up
  
  by about two orders of magnitude, and reducing the overall configure
  
  runtime by 50-70% .
  resolve() is rewritten to avoid subshells, and in unique() and
  
  flatten_extralibs() we "inline" the filter[_out] functionality.
  Note that logically, "unique" functionality has more than one possible
  
  output (depending on which of the recurring items is kept). As it
  
  turns out, other parts expect the last recurring item to be kept
  
  (which was the original behavior of uniqie()). This patch preservs
  
  its output order.
  Tested-by : Michael Niedermayer <michael@niedermayer.cc>
  
  Tested-by : Helmut K. C. Tessarek <tessarek@evermeet.cx>
  
  Tested-by : Dave Yeo <daveryeo@telus.net>
  
  Tested-by : Reino Wijnsma <rwijnsma@xs4all.nl>
  
  Signed-off-by : James Almer <jamrial@gmail.com>
  

  • [DH] configure