Recherche avancée

Sur d’autres sites (9490)

• How to force specific AVInputFormat in code (FFMPEG) ?

  18 février 2020, par kugipark

  ++Plaease understand that maybe some words or sentences could not be correct English.++

  I’m novice programmer and developing a video player app for Android which can play 5ch .avi video from http. It based on ffplay.c in FFMPEG libarary.

  Currently I have 2 problems for this below. It has occured only in android devices.

  ---

  1) Too much time taken for detecting format.

  • opening and finding stream info takes more than a minute when I trying to open 5ch video source whereas normal .mp4 (h264) source open almost immediately.

  2) Demuxing is too slow when it comes to large size video.

  • If the resolution of video getting larger, then displaying frame rate getting slower even though there are enough memory, network and CPU resources physically.

  To resolve my problem, I tried to force the format and decoders but I couldn’t found some information for specifying the input source in code level.
  The official doc site only refers to a sentence about those parameters like "If non-NULL, this parameter forces a specific input format. Otherwise the format is autodetected.". So I don’t have any clues how to set the number of streams, decoders, decoders’ private options, and etc. (And also which parameters should I manage.) If someone knows how to set the options(like AVDictionary), and pass to the av functions, please let me know an exmple. The source contains 2 video streams, 1 audio stream, and 2 more extra streams (for custom data like gps). The stream information of video is below. I printed it manually, and these are auto-detected information.

      ---------- File format information ----------
  
  flags=2097152
  
  video_codec_id=0 (NONE)
  
  audio_codec_id=0 (NONE)
  
  ctx_flags=0
  
  data_codec_id=0 (NONE)
  
  format_whitelist=(null)
  
  iformat=0xa19d0d2c
  
  ---------- Stream information ----------
  
  stream 1 of 5:
  
  ----- common ----------
  
  bit_rate: 11383235
  
  bits_per_coded_sample: 24
  
  bits_per_raw_sample: 0
  
  codec_id: 0x1C (H264)
  
  codec_tag: 875967048
  
  extradata_size: 0
  
  level: -99
  
  profile: -99
  
  sample_rate: 0
  
  ----- Video Stream ----------
  
  chroma_location: 0
  
  color_primaries: 2
  
  color_space: 2
  
  color_trc: 2
  
  field_order: 0
  
  format: -1 (NONE)
  
  height: 1080
  
  width: 1920
  
  sample_aspect_ratio.den: 1
  
  sample_aspect_ratio.num: 0
  
  video_delay: 0
  
  ----------------------------------------
  
  stream 2 of 5:
  
  ----- common ----------
  
  bit_rate: 6185438
  
  bits_per_coded_sample: 24
  
  bits_per_raw_sample: 0
  
  codec_id: 0x1C (H264)
  
  codec_tag: 875967048
  
  extradata_size: 0
  
  level: -99
  
  profile: -99
  
  sample_rate: 0
  
  ----- Video Stream ----------
  
  chroma_location: 0
  
  color_primaries: 2
  
  color_space: 2
  
  color_trc: 2
  
  field_order: 0
  
  format: -1 (NONE)
  
  height: 720
  
  width: 1280
  
  sample_aspect_ratio.den: 1
  
  sample_aspect_ratio.num: 0
  
  video_delay: 0
  
  ----------------------------------------
  
  stream 3 of 5:
  
  ----- common ----------
  
  bit_rate: 352800
  
  bits_per_coded_sample: 16
  
  bits_per_raw_sample: 0
  
  codec_id: 0x10000 (PCM_S16LE)
  
  codec_tag: 1
  
  extradata_size: 0
  
  level: -99
  
  profile: -99
  
  sample_rate: 22050
  
  ----- Audio Stream ----------
  
  block_align: 2
  
  channels: 1
  
  channel_layout: 0
  
  color_range: 0
  
  frame_size: 0
  
  initial_padding: 0
  
  seek_preroll: 0
  
  trailing_padding: 0
  
  ----------------------------------------
  
  stream 4 of 5:
  
  ----- common ----------
  
  bit_rate: 15625
  
  bits_per_coded_sample: 0
  
  bits_per_raw_sample: 0
  
  codec_id: 0x0 (NONE)
  
  codec_tag: 0
  
  extradata_size: 0
  
  level: -99
  
  profile: -99
  
  sample_rate: 0
  
  ----- Subtitle Stream ----------
  
  ----------------------------------------
  
  stream 5 of 5:
  
  ----- common ----------
  
  bit_rate: 33862
  
  bits_per_coded_sample: 0
  
  bits_per_raw_sample: 0
  
  codec_id: 0x0 (NONE)
  
  codec_tag: 0
  
  extradata_size: 0
  
  level: -99
  
  profile: -99
  
  sample_rate: 0
  
  ----- Subtitle Stream ----------

• examples : Add a VA-API encode example.

  6 novembre 2017, par Jun Zhao

  examples : Add a VA-API encode example.
  Supports only raw NV12 input.
  Example use :
  
  ./vaapi_encode 1920 1080 test.yuv test.h264
  Signed-off-by : Jun Zhao <jun.zhao@intel.com>
  
  Signed-off-by : Liu, Kaixuan <kaixuan.liu@intel.com>
  
  Signed-off-by : Mark Thompson <sw@jkqxz.net>
  

  • [DH] configure
  • [DH] doc/examples/Makefile
  • [DH] doc/examples/vaapi_encode.c

• Getting video properties with Python without calling external software

  24 juillet 2019, par ullix

  [Update :] Yes, it is possible, now some 20 months later. See Update3 below ! [/update]

  Is that really impossible ? All I could find were variants of calling FFmpeg (or other software). My current solution is shown below, but what I really would like to get for portability is a Python-only solution that doesn’t require users to install additional software.

  After all, I can easily play videos using PyQt’s Phonon, yet I can’t get simply things like dimension or duration of the video ?

  My solution uses ffmpy (http://ffmpy.readthedocs.io/en/latest/ffmpy.html ) which is a wrapper for FFmpeg and FFprobe (http://trac.ffmpeg.org/wiki/FFprobeTips). Smoother than other offerings, yet it still requires an additional FFmpeg installation.

      import ffmpy, subprocess, json
  
      ffprobe = ffmpy.FFprobe(global_options="-loglevel quiet -sexagesimal -of json -show_entries stream=width,height,duration -show_entries format=duration -select_streams v:0", inputs={"myvideo.mp4": None})
  
      print("ffprobe.cmd:", ffprobe.cmd)  # printout the resulting ffprobe shell command
  
      stdout, stderr = ffprobe.run(stderr=subprocess.PIPE, stdout=subprocess.PIPE)
  
      # std* is byte sequence, but json in Python 3.5.2 requires str
  
      ff0string = str(stdout,'utf-8')
  
  
  
      ffinfo = json.loads(ff0string)
  
      print(json.dumps(ffinfo, indent=4)) # pretty print
  
  
  
      print("Video Dimensions: {}x{}".format(ffinfo["streams"][0]["width"], ffinfo["streams"][0]["height"]))
  
      print("Streams Duration:", ffinfo["streams"][0]["duration"])
  
      print("Format Duration: ", ffinfo["format"]["duration"])

  Results in output :

      ffprobe.cmd: ffprobe -loglevel quiet -sexagesimal -of json -show_entries stream=width,height,duration -show_entries format=duration -select_streams v:0 -i myvideo.mp4
  
      {
  
          "streams": [
  
              {
  
                  "duration": "0:00:32.033333",
  
                  "width": 1920,
  
                  "height": 1080
  
              }
  
          ],
  
          "programs": [],
  
          "format": {
  
              "duration": "0:00:32.064000"
  
          }
  
      }
  
      Video Dimensions: 1920x1080
  
      Streams Duration: 0:00:32.033333
  
      Format Duration:  0:00:32.064000

  UPDATE after several days of experimentation : The hachoire solution as proposed by Nick below does work, but will give you a lot of headaches, as the hachoire responses are too unpredictable. Not my choice.

  With opencv coding couldn’t be any easier :

  import cv2
  
  vid = cv2.VideoCapture( picfilename)
  
  height = vid.get(cv2.CAP_PROP_FRAME_HEIGHT) # always 0 in Linux python3
  
  width  = vid.get(cv2.CAP_PROP_FRAME_WIDTH)  # always 0 in Linux python3
  
  print ("opencv: height:{} width:{}".format( height, width))

  The problem is that it works well on Python2 but not on Py3. Quote : "IMPORTANT NOTE : MacOS and Linux packages do not support video related functionality (not compiled with FFmpeg)" (https://pypi.python.org/pypi/opencv-python).

  On top of this it seems that opencv needs the presence of the binary packages of FFmeg at runtime (https://docs.opencv.org/3.3.1/d0/da7/videoio_overview.html).

  Well, if I need an installation of FFmpeg anyway, I can stick to my original ffmpy example shown above :-/

  Thanks for the help.

  UPDATE2 : master_q (see below) proposed MediaInfo. While this failed to work on my Linux system (see my comments), the alternative of using pymediainfo, a py wrapper to MediaInfo, did work. It is simple to use, but it takes 4 times longer than my initial ffprobe approach to obtain duration, width and height, and still needs external software, i.e. MediaInfo :

  from pymediainfo import MediaInfo
  
  media_info = MediaInfo.parse("myvideofile")
  
  for track in media_info.tracks:
  
      if track.track_type == 'Video':
  
          print("duration (millisec):", track.duration)
  
          print("width, height:", track.width, track.height)

  UPDATE3 : OpenCV is finally available for Python3, and is claimed to run on Linux, Win, and Mac ! It makes it really easy, and I verfied that external software - in particular ffmpeg - is NOT needed !

  First install OpenCV via Pip :

  pip install opencv-python

  Run in Python :

  import cv2
  
  cv2video = cv2.VideoCapture( videofilename)
  
  height = cv2video.get(cv2.CAP_PROP_FRAME_HEIGHT)
  
  width  = cv2video.get(cv2.CAP_PROP_FRAME_WIDTH) 
  
  print ("Video Dimension: height:{} width:{}".format( height, width))
  
  
  
  framecount = cv2video.get(cv2.CAP_PROP_FRAME_COUNT ) 
  
  frames_per_sec = cv2video.get(cv2.CAP_PROP_FPS)
  
  print("Video duration (sec):", framecount / frames_per_sec)
  
  
  
  # equally easy to get this info from images
  
  cv2image = cv2.imread(imagefilename, flags=cv2.IMREAD_COLOR  )
  
  height, width, channel  = cv2image.shape
  
  print ("Image Dimension: height:{} width:{}".format( height, width))

  I also needed the first frame of a video as an image, and used ffmpeg for this to save the image in the file system. This also is easier with OpenCV :

  hasFrames, cv2image = cv2video.read()   # reads 1st frame
  
  cv2.imwrite("myfilename.png", cv2image) # extension defines image type

  But even better, as I need the image only in memory for use in the PyQt5 toolkit, I can directly read the cv2-image into an Qt-image :

  bytesPerLine = 3 * width
  
  # my_qt_image = QImage(cv2image, width, height, bytesPerLine, QImage.Format_RGB888) # may give false colors!
  
  my_qt_image = QImage(cv2image.data, width, height, bytesPerLine, QImage.Format_RGB888).rgbSwapped() # correct colors on my systems

  As OpenCV is a huge program, I was concerned about timing. Turned out, OpenCV was never behind the alternatives. I takes some 100ms to read a slide, all the rest combined takes never more than 10ms.

  I tested this successfully on Ubuntu Mate 16.04, 18.04, and 19.04, and on two different installations of Windows 10 Pro. (Did not have Mac avalable). I am really delighted about OpenCV !

  You can see it in action in my SlideSorter program, which allows to sort images and videos, preserve sort order, and present as slideshow. Available here : https://sourceforge.net/projects/slidesorter/