Recherche avancée

Sur d’autres sites (4249)

• libavcodec : how to encode raw video (YUV420P) frame to the valid bitmap for saving to the ".bmp" file ?

  26 octobre 2020, par Joe Dudinski

  I'm trying to save random video frames into the '.bmp' files by encoding raw video (YUV420P) into AV_CODEC_ID_BMP format using libavcodec, but result is broken image, contains 3 RGB channels :

  


  Resulting bmp


  


  Blue channel :


  


  Green channel :


  


  Red channel :


  


  Original YUV image :


  


  My code to get current yuv frame and encode it is into BMP :

  


  int __create_bmp(AVPacket *pkt, AVCodecContext *video_dec_ctx, int video_stream_idx) {
    int err = 0;
    int got_frame = 0;
    static int vfnum = 0; //current v.frame num

    if (pkt->stream_index == video_stream_idx) {

        fprintf(stderr, "AVCodecContext codec: (name: %s), ID %d\n",
                video_dec_ctx->codec->name, video_dec_ctx->codec_id);

        /* decode video frame */
        AVFrame *frame = av_frame_alloc();

        // !!! 'avcodec_decode_video2' IS DEPRECATED !!!
        err = avcodec_decode_video2(video_dec_ctx, frame, got_frame, pkt);
        if (err < 0) {
            fprintf(stderr, "Error decoding video frame (%s)\n", av_err2str(err));
            return err;
        }
        if(got_frame == 0) return 1;

        // !!! this dbg func save mjpeg YUV420P into '.jpg' file FINE !!!
        __encode_image_frame(frame);

        AVCodec *bmpc = avcodec_find_encoder(AV_CODEC_ID_BMP);
        if(bmpc == NULL){
            fprintf(stderr, "'AV_CODEC_ID_BMP' encoder NOT FOUND!.\n");
            exit(-1);
        }
        AVCodecContext *__bmpcc = avcodec_alloc_context3(bmpc);
        __bmpcc->width = frame->width;
        __bmpcc->height = frame->height;
        __bmpcc->pix_fmt = AV_PIX_FMT_BGR24;
        __bmpcc->time_base = (AVRational){1,1};

        err = avcodec_open2(__bmpcc, bmpc, NULL);
        if(err == 0){
            fprintf(stderr, "avcodec_open2 success.\n");
        }else{
            fprintf(stderr, "avcodec_open2 ERROR!\n");
        }

        err = avcodec_send_frame(__bmpcc, frame);
        if(err == 0){
            fprintf(stderr, "avcodec_send_frame success.\n");
        }
        else{
            fprintf(stderr, "avcodec_send_frame ERROR! (code: %d)\n",err);
        }

        AVPacket *bmp_pkt = av_packet_alloc();
        err = avcodec_receive_packet(__bmpcc, bmp_pkt);
        if(err == 0){
            fprintf(stderr, "avcodec_receive_packet success.\n");
        }
        else{
            fprintf(stderr, "avcodec_receive_packet ERROR! (code: %d)\n",err);
        }
        av_frame_unref(frame);

        /* write to bitmap file */
        char outimg[2048] = {0};
        snprintf(outimg, 2048,"out-%06d.bmp", vfnum);
        FILE *outf = fopen(outimg, "wb");
        vfnum++;
        if (!outf) {
            fprintf(stderr, "Could not open destination file %s\n", outimg);
        }else{
            fwrite(bmp_pkt->data, 1, bmp_pkt->size , outf);
            fclose(outf);
        }
        av_packet_unref(bmp_pkt);
        return 0;
    }
    return 1;
}


  


  Does anyone know how to do this correctly ?

  


• FFMPEG "Could not allocate memory" Errors through php

  10 décembre 2019, par Applepiee

  I’ve tried to find a solution online for days now but cannot find any solution.

  I just switched server (Intel Xeon to AMD) and since the switch I’ve not been able to get ffmpeg conversions working through the php script. ffmpeg was the exact same version and all php settings are set.

  All commands that the script executes (copied from log files) were tried in shell and executed with no problems.

  The errors look like this :

  [124] =>       handler_name    : Video Media Handler
  
      [125] =>     Stream #8:1(eng): Audio: aac (LC) (mp4a / 0x6134706D), 48000 Hz, stereo, fltp, 160 kb/s (default)
  
      [126] =>     Metadata:
  
      [127] =>       handler_name    : Sound Media Handler
  
      [128] => Stream mapping:
  
      [129] =>   Stream #0:0 (h264) -> concat:in0:v0
  
      [130] =>   Stream #1:0 (h264) -> concat:in1:v0
  
      [131] =>   Stream #2:0 (h264) -> concat:in2:v0
  
      [132] =>   Stream #3:0 (h264) -> concat:in3:v0
  
      [133] =>   Stream #4:0 (h264) -> concat:in4:v0
  
      [134] =>   Stream #5:0 (h264) -> concat:in5:v0
  
      [135] =>   Stream #6:0 (h264) -> concat:in6:v0
  
      [136] =>   Stream #7:0 (h264) -> concat:in7:v0
  
      [137] =>   scale -> Stream #0:0 (libvpx)
  
      [138] => Press [q] to stop, [?] for help
  
      [139] => [h264 @ 0x34f1dc0] get_buffer() failed
  
      [140] => [h264 @ 0x34f1dc0] thread_get_buffer() failed
  
      [141] => [h264 @ 0x34f1dc0] decode_slice_header error
  
      [142] => [h264 @ 0x34f1dc0] no frame!
  
      [143] => [h264 @ 0x350e680] Error splitting the input into NAL units.
  
      [144] => [h264 @ 0x352af40] Cannot allocate memory.
  
      [145] => [h264 @ 0x352af40] Could not allocate memory
  
      [146] => [h264 @ 0x352af40] h264_slice_header_init() failedError while decoding stream #0:0: Cannot allocate memory
  
      [147] => [h264 @ 0x352af40] Cannot allocate memory.
  
      [148] => [h264 @ 0x352af40] Could not allocate memory
  
      [149] => [h264 @ 0x352af40] h264_slice_header_init() failedError while decoding stream #0:0: Cannot allocate memory
  
      [150] => [h264 @ 0x352af40] Cannot allocate memory.
  
      [151] => [h264 @ 0x352af40] Could not allocate memory
  
      [152] => [h264 @ 0x352af40] h264_slice_header_init() failedError while decoding stream #0:0: Cannot allocate memory
  
      [153] => [h264 @ 0x352af40] Cannot allocate memory.
  
      [154] => [h264 @ 0x352af40] Could not allocate memory
  
      [155] => [h264 @ 0x352af40] h264_slice_header_init() failedError while decoding stream #0:0: Cannot allocate memory
  
      [156] => [h264 @ 0x352af40] Cannot allocate memory.
  
      [157] => [h264 @ 0x352af40] Could not allocate memory
  
  ...
  
  ..
  
    [211519] => [h264 @ 0x886a3c0] h264_slice_header_init() failedToo many errors when draining, this is a bug. Stop draining and force EOF.
  
      [211520] => Error while decoding stream #7:0: Internal bug, should not have happened
  
      [211521] => Cannot allocate memory.
  
      [211522] => sws: initFilter failed
  
      [211523] => [Parsed_scale_1 @ 0x8e0be40] Failed to configure output pad on Parsed_scale_1
  
      [211524] => Error reinitializing filters!
  
      [211525] => Error while filtering: Operation not permitted
  
      [211526] => Finishing stream 0:0 without any data written to it.
  
      [211527] => [libvpx @ 0x6f4c600] v1.8.1-301-g89375f031
  
      [211528] => Output #0, webm, to '/home/website/public_html/media/videos/tmb/2420/video_copy.webm':
  
      [211529] =>   Metadata:
  
      [211530] =>     major_brand     : isom
  
      [211531] =>     minor_version   : 512
  
      [211532] =>     compatible_brands: isomiso2avc1mp41
  
      [211533] =>     title           : Aibeya The Animation
  
      [211534] =>     encoder         : Lavf58.35.100
  
      [211535] =>     Chapter #0:0: start 0.000000, end 102.102000
  
      [211536] =>     Metadata:
  
      [211537] =>       title           : Intro
  
      [211538] =>     Chapter #0:1: start 102.102000, end 110.369000
  
      [211539] =>     Metadata:
  
      [211540] =>       title           : Title
  
      [211541] =>     Chapter #0:2: start 110.369000, end 312.179000
  
      [211542] =>     Metadata:
  
      [211543] =>       title           : Part 1
  
      [211544] =>     Chapter #0:3: start 312.179000, end 548.415000
  
      [211545] =>     Metadata:
  
      [211546] =>       title           : Part 2
  
      [211547] =>     Chapter #0:4: start 548.415000, end 706.831000
  
      [211548] =>     Metadata:
  
      [211549] =>       title           : Part 3
  
      [211550] =>     Chapter #0:5: start 706.831000, end 1011.052000
  
      [211551] =>     Metadata:
  
      [211552] =>       title           : Part 4
  
      [211553] =>     Chapter #0:6: start 1011.052000, end 1198.823000
  
      [211554] =>     Metadata:
  
      [211555] =>       title           : Part 5
  
      [211556] =>     Chapter #0:7: start 1198.823000, end 1501.408000
  
      [211557] =>     Metadata:
  
      [211558] =>       title           : Part 6
  
      [211559] =>     Chapter #0:8: start 1501.408000, end 1579.945000
  
      [211560] =>     Metadata:
  
      [211561] =>       title           : Part 7
  
      [211562] =>     Chapter #0:9: start 1579.945000, end 1654.293000
  
      [211563] =>     Metadata:
  
      [211564] =>       title           : Ending
  
      [211565] =>     Stream #0:0: Video: vp8 (libvpx), yuv420p, 400x240 [SAR 837:785 DAR 279:157], q=10-42, 250 kb/s, 23.98 fps, 1k tbn, 23.98 tbc (default)
  
      [211566] =>     Metadata:
  
      [211567] =>       encoder         : Lavc58.64.101 libvpx
  
      [211568] =>     Side data:
  
      [211569] =>       cpb: bitrate max/min/avg: 0/0/0 buffer size: 600000 vbv_delay: N/A
  
      [211570] => frame=    0 fps=0.0 q=0.0 Lsize=       1kB time=00:00:00.00 bitrate=N/A speed=   0x
  
      [211571] => video:0kB audio:0kB subtitle:0kB other streams:0kB global headers:0kB muxing overhead: unknown
  
      [211572] => Output file is empty, nothing was encoded (check -ss / -t / -frames parameters if used)
  
      [211573] => Conversion failed!
  
  )

  Command example used :

  ffmpeg_command is /usr/local/bin/ffmpeg -ss 100 -t 1 -i /home/website/public_html/media/videos/iphone/2407.mp4 -ss 197 -t 2 -i /home/website/public_html/media/videos/iphone/2407.mp4 -ss 294 -t 3 -i /home/website/public_html/media/videos/iphone/2407.mp4 -ss 391 -t 3 -i /home/website/public_html/media/videos/iphone/2407.mp4 -ss 488 -t 3 -i /home/website/public_html/media/videos/iphone/2407.mp4 -ss 585 -t 3 -i /home/website/public_html/media/videos/iphone/2407.mp4 -ss 682 -t 3 -i /home/website/public_html/media/videos/iphone/2407.mp4 -ss 779 -t 3 -i /home/website/public_html/media/videos/iphone/2407.mp4 -ss 876 -t 3 -i /home/website/public_html/media/videos/iphone/2407.mp4 -filter_complex "[0][1][2][3][4][5][6][7]concat=n=8:v=1:a=0",scale=400:240 -codec:v libx264 -unsharp -b:v 250k -maxrate 250k -bufsize 600k -qmin 10 -qmax 42 -threads 4 -an -y /home/website/public_html/media/videos/tmb/2407/video_copy.mp4

  PHP Info :
  PHP 5.6
  memory_limit 2001M
  max_execution_time 7200
  upload_max_filesize 2000M
  post_max_size 2000M
  max_input_time 7200
  exec is not disabled

  Server info
  CentOS Linux 7 (Core)
  ADVANCE-4 - AMD Epyc 7351P - 128GB DDR4 ECC 2400MHz - 2x HDD SATA 4TB Datacenter Class + 2x SSD NVMe 500GB Enterprise Class Soft RAID

  All help is really appreciated ! Thanks in advance

• "Smooth" the ffmpeg showwaves video filter output

  24 décembre 2019, par goodytx

  I think I’m close to what I’m after, thanks to excellent answers like : FFMPEG : Fill/Change (part of) audio waveform color as per actual progress with respect to time progress ...but not quite there !

  I’m scripting the generation of a waveform video with showwaves, and overlaying that onto a static image background, turning the whole thing into video. All is well, except I’d like to "smooth" the waveform. In most tutorials/answers you will see jagged "peaks" in the waveform animation, as of course the audio data varies widely into peaks and troughs. (The differences expressed between the lowest and highest amplitude (I think) in each audio frame sample.) I’d like to get more of a consistent "flowing wave", without the sharp spikes up and down.

  Currently using the filter much as described in the previous link :

  [1:a]aformat=channel_layouts=mono,asplit[red][white]; \
  
  [red]showwaves=s=1280x100:rate=15:mode=cline:scale=sqrt:colors=0xff0000[red]; \

  Which results in the following (ignoring the red on white overlay, which I may drop) :

  Current v desired

  This generates the type of wave plot I want (solid vs a sine, for example), but I think I need to either tweak the audio stream or...possibly post filter the showwaves video stream (but I think the former) to get the effect I want. I tried to reduce the rate of showwaves but that just results in juddery video. So to me that feels like the audio data needs to be the changed factor.

  My limited audio knowledge leads me to think I should gain, compress, normalize, or all of the above, the audio input, which I have tried, but can’t seem to nail it. Keep in mind the audio stream in this filter chain is purely feeding the showwaves filter, so I can "trash" it as much as needed in order to get the visualization I want.

  I hope this all makes sense, and if I need to provide more info, please let me know. Thanks !