Recherche avancée

Sur d’autres sites (11936)

• avcodec/adpcm_ms : support custom block size for encoding

  14 octobre 2020, par Zane van Iperen

  avcodec/adpcm_ms : support custom block size for encoding
  Fixes tickets #6585 and #7109
  Signed-off-by : Zane van Iperen <zane@zanevaniperen.com>
  

  • [DH] libavcodec/adpcmenc.c

• Do ffmpeg create thumbnail download all of file via URL nodejs ?

  11 novembre 2019, par Nam Nguyễn

  I am creating a thumbnail from video URL by ffmpeg.

  This is my code :

  function createThumbnail(videoURL, keyFileThumb) {
  
      return new Promise((resolve, reject) => {
  
          let tmpFile = createWriteStream(`./${keyFileThumb}`)
  
          const ffmpeg = spawn(ffmpegPath, [
  
              '-ss',
  
              '00:00:01.000',
  
              '-i',
  
              videoURL,
  
              '-frames:v',
  
              '1',
  
              '-f',
  
              'image2',
  
              '-c:v',
  
              'mjpeg',
  
              'pipe:1'
  
          ])
  
  
  
          ffmpeg.stdout.pipe(tmpFile)
  
  
  
          ffmpeg.on('close', function (code) {
  
              tmpFile.end()
  
              resolve(code)
  
          })
  
  
  
          ffmpeg.on('error', function (err) {
  
              console.log(err)
  
              reject(err)
  
          })
  
      })
  
  }

  I just want to create thumbnail from stream video URL then disconnect. I do not want download all of file to generate thumbnail in few second of video.

• Making changes on a video before a download [on hold]

  1er février 2016, par Chris Russo

  We have a website with lots of videos (>10,000) and lots of users (>50,000). Each video might be around 1,2 Gbs.

  The goal is to generate a unique video for each one of the members or users of the website making a small variation either on the audio track of the video, or the keyframes on the video itself, using low frequency beeps, or a few random pixels on specific keyframes.

  I understand that this operation can be done using SoX for the audio track, or FFmpeg in the case of the video, manipulating the keyframes.

  Here’s my question, as we have to choose between 2 options at this point :

  In terms of resources : would it be convenient or even possible to generate in real time a clone of the video and make this variations at the moment that the user’s request the download (please considerate the big volume of videos and users).

  Or it would be convenient to face this problem using HDD, pre-generating and storing the already manipulated videos ready to be served ?

  I understand that storing all this data will require a few many Tera’s but I’m not sure about how much RAM and Microprocessing could be demanded to perform this operations in real time, over videos that might be as big as 1Gbs.