Recherche avancée

Sur d’autres sites (9238)

• FFmpeg Muxing binary data with video into MPEG-TS stream

  22 avril 2020, par diogoaos

  I'm trying to create a MPEG-TS stream with FFmpeg. I feed FFmpeg video form a file and binary data from a UDP stream.

  



  ffmpeg -re -i video.mp4 \
       -f data -i udp://localhost:5000 \
       -map 0:0 -map 0:1 -map 1:0 -codec copy \
       -f mpegts test.ts


  



  I use socat to connect to FFmpeg and type random data :

  



  socat udp:localhost:5000 -


  



  When I try to demux the data channel (channel 2 is for data), it's empty :

  



  ffmpeg -i test.ts -map 0:2 -c:d copy -f data -


  



  I've also tried doing this feeding directly text files and that works fine (I can demux the data stream from the resulting .ts file and it's equal). I've also tried using named pipes connected to a Python script, but that didn't work well (FFmpeg seems to wait for an EOF and than does not keep reading the named pipe).

  



  How do I mux video and binary data from different sources into a single MPEG Transport Stream ?

  


• FFMPEG - local video to UDP streaming to OpenCV - video quality degraded

  6 juin 2021, par user3925023

  my goal is to re-stream local video content / desktop screencasting, to an UDP flow that I need to process on a Python script.

  



  Here is the FFMPEG script that I'm using :

  



  ffmpeg -re -i C:\Users\test\Downloads\out.ts -strict -2 -c:v copy -an -preset slower -tune stillimage -b 11200k -f rawvideo udp://127.0.0.1:5000


  



  And here is the simple Python script supposed to read the stream flow :

  



  import cv2

cap = cv2.VideoCapture('udp://127.0.0.1:5000',cv2.CAP_FFMPEG)
if not cap.isOpened():
    print('VideoCapture not opened')
    exit(-1)
width = cap.get(cv2.CAP_PROP_FRAME_WIDTH)  # float
height = cap.get(cv2.CAP_PROP_FRAME_HEIGHT)  # float
print(str(width))
print(str(height))
while True:
    ret, frame = cap.read()
    imgray = frame[int(round((height/100)*70,0)):int(round((width/100)*42,0)), int(round((height/100)*74,0)):int(round((width/100)*54,0))]
    if not ret:
        print('frame empty')
        break
    cv2.imshow('image', imgray)
    if cv2.waitKey(1)&0XFF == ord('q'):
        break
cap.release()


  



  I'm able to visualize portion of the stream video as expect, but I'm facing lot of issue in video quality degradation, specially video artifact probably due missing packet processing :

  



  

  



  Also these are error log I'm geting from script :

  



  [h264 @ 0000026eb272f280] error while decoding MB 105 66, bytestream -21
[h264 @ 0000026eb2fcb740] error while decoding MB 100 53, bytestream -11
[h264 @ 0000026eb272f280] error while decoding MB 32 22, bytestream -11
[h264 @ 0000026ead9ee300] error while decoding MB 60 20, bytestream -25
[h264 @ 0000026eb27f00c0] error while decoding MB 9 62, bytestream -5
[h264 @ 0000026ead9ee780] error while decoding MB 85 44, bytestream -5
[h264 @ 0000026eb27f0800] error while decoding MB 64 25, bytestream -15
[h264 @ 0000026eb272f280] error while decoding MB 112 23, bytestream -17
[h264 @ 0000026eb2735200] error while decoding MB 30 21, bytestream -7


  



  Actually I don't care about video fluidity,I can also reduce the FPS, important thing is the video quality. Not sure if I'm doing wrong on the scripting python part or if I'm using wrong FFMPEG command.

  



  Many Thanks

  


• Combining 2 FFMPEG Commands

  11 juillet 2020, par John Doe

  I'm trying to combine 2 ffmpeg commands, one which creates the video, and another which adds a simple fade to the beginning of the created video. Here's what I have :

  


  ffmpeg -y -stream_loop -1 -i "video.mp4" -stream_loop -1 -i "music.mp3" -i "audio.mp3" -filter_complex "[1:a]volume=0.1[a1];[2:a]adelay=5000|5000,apad=pad_dur=10[a2];[a1][a2]amerge=inputs=2,afade=in:st=0:d=5[audio]" -map "0:v" -map "[audio]" -c:v libx264 -c:a aac -ac 2 -ar 22050 -preset veryfast -shortest "output.mp4"

ffmpeg -y -i "output.mp4" -filter_complex "[0:v]fade=in:0:d=5" -c:a copy -preset veryfast -movflags faststart -fflags genpts "done.mp4"


  


  The two commands work perfectly fine, however the second one takes about the same amount of time to process as the first, and I feel it should be relatively easy to do the fade-in during the first encode. For my skillset atleast, I was wrong. Please could someone with more experience lend a helping hand ?

  


  Thanks.