Recherche avancée

Sur d’autres sites (13619)

• path issues with FFMPEG Bash script to concat and encode across multiple subfolders

  27 décembre 2022, par NoobCoder

  I'm trying to write a bash script for Mac OSx Terminal to compress a series of GoPro .MP4 videos from the SDcard directly into a smaller .MP4s on a local network server. The GoPro saves .MP4s in the 100GOPRO folder on the card. After filming, I will through that folder and manually put .MP4s from each game into subfolders within the 100GOPRO folder, named A1, A2, A3, etc.

  


  Folder structure

  


  /GoPro/DCIM/100GOPRO/
               -------/A1/
                       -----GX01xxx1.mp4
                       -----GX01xxx2.mp4
               -------/A2/
                       -----GX01xxx3.mp4
                       -----GX01xxx4.mp4
                       -----GX01xxx5.mp4
                       -----GX01xxx6.mp4


  


  ...etc

  


  I would like then like to run a script from the 100GOPRO folder that will do these steps :

  


  

  1. Within each subfolder, auto-create a file.txt with the names of the subfolder's .MP4s in the format to concat the files (each line has "file 'GX01xxx3.mp4'")

  


  2. Pass that subfolder's file.txt as the input to ffmpeg to reencode and save to a network folder with the name A1.mp4 or A2.mp4

  


  3. Repeat for each subfolder and quit.

  


  


  I'm getting hung up on the dynamic path to the subfolder's file.txt. My code just creates a file.txt in the 100GOPRO folder, and appends all the subfolder contents into that single long combined text file. The output then would create a correct first MP4, but second MP4 contains folder 1 and 2, then 3 contains 1, 2, and 3, etc.

  


  Here's the script I ran :

  


  #!/bin/bash
for f in A*/*.mp4 ; do
echo file \'$f\' >> list.txt ;
done && ffmpeg -f concat -safe 0 -i list.txt /Volume/Server/Videos/A$f.mp4 && rm list.txt


  


  Clearly, failing in how that path for echo to save in the subfolder A*, how to call that subfolder's file.txt as the input for ffmpeg, and how to name the output after the folder.

  


  Thanks for any help you can offer.

  


• C++ h264 ffmpeg/libav encode/decode(lossless) issues

  1er février 2017, par MrSmith

  Insights to encode/decode video with ffmpeg h264 (lossless)

  So I got something working on the encoding part, encode an avi in 264 however VLC wont play it, however Totem will.
  Decoding the same file proves troublesome. (I want the exact same data/frame going in as going out), I get these ;

  saving frame   5
  
  Video decoding
  
  [h264 @ 0x1d19880] decode_slice_header error
  
  frame :6
  
  saving frame   6
  
  Video decoding
  
  [h264 @ 0x1d19880] error while decoding MB 15 7, bytestream -27
  
  [h264 @ 0x1d19880] concealing 194 DC, 194 AC, 194 MV errors in I frame
  
  frame :7
  
  saving frame   7
  
  Video decoding
  
  [h264 @ 0x1d19880] decode_slice_header error

  and ultimatly this

  [H264 Decoder @ 0x7f1320766040] frame :11
  
  Broken frame packetizing
  
  [h264 @ 0x1d19880] SPS changed in the middle of the frame
  
  [h264 @ 0x1d19880] decode_slice_header error
  
  [h264 @ 0x1d19880] no frame!
  
  Error while decoding frame 11

  GAME OVER.

  Now I suspect that I have to go back to 1. the encoding part, there is problary a good reason VLC wont play it !

  I encode like this.

  void encode(char *Y,char *U,char *V){
  
  av_init_packet(&pkt);
  
  pkt.data = NULL;    // packet data will be allocated by the encoder
  
  pkt.size = 0;
  
  fflush(stdout);
  
  
  
  frame->data[0] = (uint8_t*)Y;
  
  frame->data[1] = (uint8_t*)U;
  
  frame->data[2] = (uint8_t*)V;
  
  frame->pts = ++i;
  
  
  
  ret = avcodec_encode_video2(c, &pkt, frame, &got_output);
  
  if (ret < 0) {
  
      fprintf(stderr, "Error encoding frame\n");
  
      exit (EXIT_FAILURE);
  
  }
  
  if (got_output) {
  
      printf("Write frame %3d (size=%5d)\n", i, pkt.size);
  
      fwrite(pkt.data, 1, pkt.size, f);
  
      av_free_packet(&pkt);
  
  }
  
  }

  And the codec is setup like this :

  AVCodecID dasd = AV_CODEC_ID_H264;
  
  codec = avcodec_find_encoder(dasd);
  
  c = avcodec_alloc_context3(codec);
  
  c->bit_rate = 400000;
  
  c->width = 320;
  
  c->height = 240;
  
  c->time_base= (AVRational){1,25};
  
  c->gop_size = 10; 
  
  c->max_b_frames=1;
  
  c->pix_fmt = AV_PIX_FMT_YUV420P;
  
  av_opt_set(c->priv_data, "preset", "slow", 0);
  
  avcodec_open2(c, codec, NULL);

  Since I am going for lossless i am not dealing with delayed frames(is this a correct assumption ?)
  I may not actually be encoding lossless, it seems like I may have to go with something like

  AVDictionary *param;
  
  av_dict_set(&param, "qp", "0", 0);

  And then open...

  So I guess me questions is these :

  • What are the correct codec params for lossless encoding (and advice if h264 is a terrible idea in this regard).
  • Do I need to handle delayed frames when going for lossless ?
  • Why is VLC mad at me ?

  Thanks.

• Why am I having issues re-encoding rtsp stream to rtmp via FFMPEG-python ? [closed]

  27 août 2023, par Luke Lopez

  I can see that the python script is not throwing errors however I cant stream the video via VLC player.

  


  
  Metadata:
    title           : Media Server
    encoder         : Lavf60.3.100
  Stream #0:0: Video: h264 ([7][0][0][0] / 0x0007), yuv420p(tv, progressive), 640x360, q=2-31, 1 kb/s, 25 fps, 1k tbn
    Metadata:
      encoder         : Lavc60.3.100 libx264
    Side data:
      cpb: bitrate max/min/avg: 0/0/1000 buffer size: 0 vbv_delay: N/A
frame= 1927 fps= 25 q=69.0 size=      92kB time=00:01:17.00 bitrate=   9.7kbits/s dup=13 drop=1 speed=0.98x     



import subprocess

# Input RTSP stream URL
input_stream_url = "rtsp://myStream"

# Wowza RTMP or RTSP URL (replace with your Wowza server's details)
wowza_url = "rtmp://myWowzaServer"

# Resolution 
desired_width = 640
desired_height = 360

# Video bitrate and frame rate settings
video_bitrate = "1240"  # 1240 Kbps
fps = 25

# Start FFmpeg process to pull and push the stream with specified settings
ffmpeg_command = [
    'ffmpeg',
    '-rtsp_transport', 'tcp',  # Use TCP for RTSP transport (optional)
    '-i', input_stream_url,    # Input RTSP stream URL
    '-s', f'{desired_width}x{desired_height}',  # Desired resolution
    '-c:v', 'libx264',        # Video codec: H.264
    '-b:v', video_bitrate,    # Video bitrate
    '-r', str(fps),            # Frame rate
    '-c:a', 'aac',            # Audio codec: AAC
    '-f', 'flv',              # Output format (FLV for RTMP, rtsp for RTSP)
    wowza_url                 # Wowza RTMP or RTSP URL
]

# Start FFmpeg as a subprocess
ffmpeg_process = subprocess.Popen(ffmpeg_command)

# Stop Condition
print("Press Enter to stop...")
input()

# Terminate FFmpeg when done
ffmpeg_process.terminate()



  


  So far I have tried implementing best encoding practices as recommended by Wowza, have been messing with the frame rate, bitrate etc... but with no luck. Hopefully someone experienced with using FFMpeg can give suggestions / solutions ??

  


  Thanks !