Recherche avancée

Sur d’autres sites (9728)

• Greed is Good ; Greed Works

  25 novembre 2010, par Multimedia Mike — VP8

  Greed, for lack of a better word, is good ; Greed works. Well, most of the time. Maybe.

  Picking Prediction Modes
  VP8 uses one of 4 prediction modes to predict a 16x16 luma block or 8x8 chroma block before processing it (for luma, a block can also be broken into 16 4x4 blocks for individual prediction using even more modes).

  So, how to pick the best predictor mode ? I had no idea when I started writing my VP8 encoder. I did not read any literature on the matter ; I just sat down and thought of a brute-force approach. According to the comments in my code :

  // naive, greedy algorithm :
  //   residual = source - predictor
  //   mean = mean(residual)
  //   residual -= mean
  //   find the max diff between the mean and the residual
  // the thinking is that, post-prediction, the best block will
  // be comprised of similar samples
  

  After removing the predictor from the macroblock, individual 4x4 subblocks are put through a forward DCT and quantized. Optimal compression in this scenario results when all samples are the same since only the DC coefficient will be non-zero. Failing that, when the input samples are at least similar to each other, few of the AC coefficients will be non-zero, which helps compression. When the samples are all over the scale, there aren’t a whole lot of non-zero coefficients unless you crank up the quantizer, which results in poor quality in the reconstructed subblocks.

  Thus, my goal was to pick a prediction mode that, when applied to the input block, resulted in a residual in which each element would feature the least deviation from the mean of the residual (relative to other prediction choices).

  Greedy Approach
  I realized that this algorithm falls into the broad general category of "greedy" algorithms— one that makes locally optimal decisions at each stage. There are most likely smarter algorithms. But this one was good enough for making an encoder that just barely works.

  Compression Results
  I checked the total file compression size on my usual 640x360 Big Buck Bunny logo image while forcing prediction modes vs. using my greedy prediction picking algorithm. In this very simple test, DC-only actually resulted in slightly better compression than the greedy algorithm (which says nothing about overall quality).

  prediction mode	quantizer index = 0 (minimum)	quantizer index = 10
  greedy	286260	98028
  DC	280593	95378
  vertical	297206	105316
  horizontal	295357	104185
  TrueMotion	311660	113480

  As another data point, in both quantizer cases, my greedy algorithm selected a healthy mix of prediction modes :

  • quantizer index 0 : DC = 521, VERT = 151, HORIZ = 183, TM = 65
  • quantizer index 10 : DC = 486, VERT = 167, HORIZ = 190, TM = 77

  Size vs. Quality
  Again, note that this ad-hoc test only measures one property (a highly objective one)— compression size. It did not account for quality which is a far more controversial topic that I have yet to wade into.

• What Tee pseudo-muxer command line for multiple output ffmpeg works for output in mp4 and stream udp unicast ?

  10 mars 2017, par Mohammad Taha

  I am using ffmpeg command line for getting multiple outputs with tee pseudo-muxer. Commands I have tried :

  Case 1- To muxed Video (video.h264) with Audio (audio.aac) in mp4 and send mpegts to udp unicast address :

  ffmpeg -i 'video.h264' -i "audio.aac" -vcodec copy -acodec copy -absf aac_adtstoasc -f tee -map 1:v -map 0:a "result.mp4|[f=mpegts]udp://127.0.0.1:2211"

  Output 1 : Resultant mp4 file has no audio and the udp stream when played is also having no audio.

  Case 2- To muxed Video (video.h264) with Audio (audio.aac) in mpegts file and send mpegts to udp unicast address :

  ffmpeg -i 'video.h264' -i "audio.aac" -vcodec copy -acodec copy -f tee -map 1:v -map 0:a "[f=mpefts]result.mpeg|[f=mpegts]udp://127.0.0.1:2211"

  Output 2 : Resultant mpegts file is fine and the udp stream when played is also having audio. NO ISSUE in this.

  I have to do Case 1 where i can mux aac and h264 and get two output one is mp4 file dump and second is stream mpegts udp unicast (like udp ://127.0.0.1:2211). And both the output should have video+audio.

  Can anybody tell me why in Case 1 the output file is not having audio ?

• Bash script to automate FFmpeg operations fails when calling the command, but copy-pasting the generated command into the terminal works [duplicate]

  28 février, par GaboScharff99

  I wrote a bash script which automates a number of conversion operations on video files using FFmpeg. Oddly enough, the FFmpeg call itself now fails when running the script, with a very confusing error message, I might add, but when I copy the command generated by the script into the terminal and run it, it works flawlessly. I'm sorry to insert such a long code block here, but considering how strange this error is, it might be anywhere in the script, so here it is :

  


  #!/bin/bash

audioTrack=1
subSource=1
subTrack=0
transcodeVideo=1
transcodeAudio=1
volumeMultiplier=1
degradeToStereo=0
subLanguage="Japanese"

while getopts "t:ns:vam:dl:h" opt; do
    case "$opt" in
        t) audioTrack=${OPTARG};;
        n) subSource=0;;
        s) subTrack=${OPTARG};;
        v) transcodeVideo=0;;
        a) transcodeAudio=0;;
        m) volumeMultiplier=${OPTARG};;
        d) degradeToStereo=1;;
        l) subLanguage=${OPTARG};;
        h)
            echo "Options:"
            echo "-t [integer]: Audio track number. Default: 1."
            echo "-n: If included, subtitles will be taken from internal source."
            echo "-s [integer]: Subtitles track number. Default: 0."
            echo "-v: If included, video source will be copied without transcoding."
            echo "-a: If included, audio source will be copied without transcoding."
            echo "-m [number]: Volume multiplier. If 1, volume is unaffected. Default: 1"
            echo "-d: If included, audio will be degraded to stereo."
            echo "-l [language]: Subtitles language. Only used for external subtitles source. Default: Japanese."
            exit 0
        ;;
    esac
done

echo "Audio track: $audioTrack."
echo "Subtitles track: $subTrack."
params="-map 0:0 -map 0:$audioTrack -map $subSource:$subTrack -c:v"

if [[ $transcodeVideo -eq 1 ]]; then
    echo "Video will be transcoded."
    params="$params hevc"
elif [[ $transcodeVideo -eq 0 ]]; then
    echo "Video will be copied without transcoding."
    params="$params copy"
fi

params="$params -c:a"

if [[ $transcodeAudio -eq 1 ]]; then
    echo "Audio will be transcoded."
    params="$params libopus"
elif [[ $transcodeAudio -eq 0 ]]; then
    echo "Audio will be copied without transcoding."
    params="$params copy"
fi

if [[ $volumeMultiplier -ne 1 ]]; then
    echo "Volume will be multiplied by a factor of $volumeMultiplier."
    params="$params -filter:a 'volume=$volumeMultiplier'"
else
    echo "Volume will be unaffected."
fi

if [[ $degradeToStereo -eq 1 ]]; then
    echo "Audio will be degraded to stereo."
    params="$params -ac 2"
elif [[ $degradeToStereo -eq 0 ]]; then
    echo "Audio will not be degraded to stereo."
fi

params="$params -c:s copy"

if [[ $subSource -eq 1 ]]; then
    echo "Subtitles source is external."
    echo "Subtitles language is $subLanguage."
    params="$params -metadata:s:s:0 title='' -metadata:s:s:0 language='$subLanguage'"
else
    echo "Subtitles source is internal."
fi

if [[ -f titles.txt ]]; then
    echo "A titles.txt file was found. Titles will be changed according to it."
    echo "Please check titles.txt to make sure the titles are correct."
    changeTitles=1
    counter=0
else
    echo "A titles.txt file was not found. Titles will not be changed."
    changeTitles=0
fi

read -p "Are these options correct? (y/n) " choice

case "$choice" in
    y|Y)
        echo "Initiating conversion sequence. This may take a while..."

        mkdir output
        currentParams=""

        shopt -s nullglob
        for i in *.mp4 *.mkv; do
            currentParams=$params
            fileNameNoExtension=$(echo $i | rev | cut -f 2- -d '.' | rev)

            if [[ $subSource -eq 1 ]]; then
                currentParams="-f srt -i $fileNameNoExtension.srt $currentParams"
            fi

            if [[ $changeTitles -eq 1 ]]; then
                ((counter++))
                currentParams="$currentParams -metadata title='$(awk "NR==$counter" titles.txt)'"
            fi

            ffmpeg -i "$i" $currentParams "output/$fileNameNoExtension.mkv"
        done

        echo "Conversion finished!"
    ;;
    n|N) echo "Operation canceled. Exiting.";;
    *) echo "Invalid input. Try again.";;
esac


  


  The directory I'm running this in contains six video files :

  


  

  1. E1 - The Pirates of Orion.mkv

  


  2. E2 - Bem.mkv

  


  3. E3 - The Practical Joker.mkv

  


  4. E4 - Albatross.mkv

  


  5. E5 - How Sharper Than a Serpent's Tooth.mkv

  


  6. E6 - The Counter-Clock Incident.mkv

  


  


  Here's the titles.txt file, for completion's sake :

  


  Star Trek: The Animated Series - Season 2, Episode 1 - The Pirates of Orion
Star Trek: The Animated Series - Season 2, Episode 2 - Bem
Star Trek: The Animated Series - Season 2, Episode 3 - The Practical Joker
Star Trek: The Animated Series - Season 2, Episode 4 - Albatross
Star Trek: The Animated Series - Season 2, Episode 5 - How Sharper Than a Serpent's Tooth
Star Trek: The Animated Series - Season 2, Episode 6 - The Counter-Clock Incident


  


  And finally, here's the error message given by FFmpeg on the terminal for every video file when running the command :

  


  Unable to find a suitable output format for 'Trek:'
Trek:: Invalid argument


  


  Maybe there are better ways to handle all of this, but first and foremost, I would like to figure out why the command fails with such a confusing error message. The only place where the string 'Trek :' is found is in the title taken from titles.txt, but I don't understand why that's seemingly being passed to the name of the output file instead of the title, and apparently only when running the script.

  


  Thanks a lot for your answers ! I know this is quite a bit of text, so I really appreciate you taking your time to read through this.