Recherche avancée

Sur d’autres sites (5679)

• I have a m3u8 file where the individual files don't have any .ts format, Is there a way to cocnat them to a single mp4 file

  6 septembre 2020, par Suhail Hussain

  Here is a snippet of the m3u8 file

  


  #EXTM3U
#EXTINF:1,0
0
#EXTINF:1695,0c9c3bf590e32dcb8c4b83222056838b
0c9c3bf590e32dcb8c4b83222056838b
#EXTINF:1,1
1
#EXTINF:4,2
2
#EXTINF:3,3
3
#EXTINF:4,4
4
#EXTINF:3,5
5
#EXTINF:3,6
6
#EXTINF:4,7
7
#EXTINF:4,8
8
#EXTINF:3,9
9
#EXTINF:4,10
10


  


  This goes on for some 500 files. I am able to open the folder in vlc as a playlist but it is just a collection of 500 files that play one after the another. I checked online and found that ffmpeg can concatenate a m3u8 file to a mp4. That unfortunately did not work. After trying a few different syntaxes that I found on different forums which also did not work, I tried "ffplay" on the file name which once again gave the same error message as before - Invalid data found when processing input:=    0B f=0/0

  


  So this made me believe perhaps ffmpeg is unable to open the file while vlc is able to. Any way to combine these files to a single file is appreciated

  


• How do I toggle individual codec options in libavcodec (specifically h264_options)

  2 juillet 2020, par John Allard

  I'm trying to figure out how to enable enable_er option as is defined in h264dec.c in libavcodec. This is defined as an AVOption as part of the AVCodec.priv_class.option field. I can't figure out if this is some sort of compile-time option or if it's an option that I can enable via the av_dict_set method when initializing an AVCodec via avcodec_open2.

  


  I'm talking about these options in h264dec.c

  


  #define OFFSET(x) offsetof(H264Context, x)
#define VD AV_OPT_FLAG_VIDEO_PARAM | AV_OPT_FLAG_DECODING_PARAM
static const AVOption h264_options[] = {
    { "is_avc", "is avc", OFFSET(is_avc), AV_OPT_TYPE_BOOL, {.i64 = 0}, 0, 1, 0 },
    { "nal_length_size", "nal_length_size", OFFSET(nal_length_size), AV_OPT_TYPE_INT, {.i64 = 0}, 0, 4, 0 },
    { "enable_er", "Enable error resilience on damaged frames (unsafe)", OFFSET(enable_er), AV_OPT_TYPE_BOOL, { .i64 = -1 }, -1, 1, VD },
    { NULL },
};

static const AVClass h264_class = {
    .class_name = "H264 Decoder",
    .item_name  = av_default_item_name,
    .option     = h264_options,
    .version    = LIBAVUTIL_VERSION_INT,
};

AVCodec ff_h264_decoder = {
    .name                  = "h264",
    .long_name             = NULL_IF_CONFIG_SMALL("H.264 / AVC / MPEG-4 AVC / MPEG-4 part 10"),
    .type                  = AVMEDIA_TYPE_VIDEO,
    .id                    = AV_CODEC_ID_H264,
    .priv_data_size        = sizeof(H264Context),
    .init                  = h264_decode_init,
    .close                 = h264_decode_end,
    .decode                = h264_decode_frame,
    .capabilities          = /*AV_CODEC_CAP_DRAW_HORIZ_BAND |*/ AV_CODEC_CAP_DR1 |
                             AV_CODEC_CAP_DELAY | AV_CODEC_CAP_SLICE_THREADS |
                             AV_CODEC_CAP_FRAME_THREADS,
    .caps_internal         = FF_CODEC_CAP_INIT_THREADSAFE | FF_CODEC_CAP_EXPORTS_CROPPING,
    .flush                 = flush_dpb,
    .init_thread_copy      = ONLY_IF_THREADS_ENABLED(decode_init_thread_copy),
    .update_thread_context = ONLY_IF_THREADS_ENABLED(ff_h264_update_thread_context),
    .profiles              = NULL_IF_CONFIG_SMALL(ff_h264_profiles),
    .priv_class            = &h264_class,
};


  


• Shows improper/corrupted TS segments from Opencv webcam and FFmpeg

  30 juin 2020, par playmaker420

  Im experimenting with opencv and ffmpeg to create a live hls stream from the webcam using some scripts

  


  The ffmpeg version i use is 3.4

  


  frame-detection.py

  


  import numpy as np
import cv2
import sys


cap = cv2.VideoCapture(0)

while(True):
    # Capture frame-by-frame
    ret, frame = cap.read()
    framestring = frame.tostring()
    sys.stdout.write(str(framestring))

    # Display the resulting frame
    cv2.imshow('frame', frame)
    if cv2.waitKey(1) & 0xFF == ord('q'):
        break

# When everything done, release the capture
cap.release()
cv2.destroyAllWindows()


  


  hlslive_generator.sh

  


  #!/bin/bash

# Create folder from args
mkdir -p $1

# Get into the folder
cd $1

# Start running FFmpeg HLS Live streaming
python frame-detection.py | ffmpeg  \
  -f rawvideo \
  -framerate 10 \
  -video_size 640x480 \
  -i - foo.mp4  \
  -vcodec libx264 \
  -acodec copy \
  -pix_fmt yuv420p \
  -color_range 2 \
  -hls_time 1 \
  -hls_list_size 5 \
  -hls_flags delete_segments \
  -use_localtime 1 \
  -hls_segment_filename '%Y%m%d-%s.ts' \
  ./playlist.m3u8


  


  I used the following commands to run the scripts and it creates a folder and generate ts segments in it

  


  ./hlslive_generator.sh hlssegments


  


  The issue i face here is with the created ts files, on playing these segments with the video player it shows improper/corrupted segments.

  


  Can someone help me to identify the issue ? Thanks in advance