Recherche avancée

Sur d’autres sites (7010)

• C++ ffmpeg Queue input is backward in time while encoding

  15 août 2022, par Turgut

  I've made a program that takes a video as an input, decodes it's video and audio data, then edits the video data and encodes both video and audio (audio remains unedited). I've managed to successfully get the edited video as an output so far, but when I add in the audio, I get an error that says Queue input is backward in time. I used the muxing example from ffmpegs doc/examples for encoding, here is what it looks like (I'm not including the video encoding parts since its working just fine) :

  


  typedef struct {
            OutputStream video_st, audio_st;
            const AVOutputFormat *fmt;
            AVFormatContext *oc;
            int have_video, have_audio, encode_video, encode_audio;
            std::string name;
        } encode_info;

encode_info enc_inf;

void video_encoder::open_audio(AVFormatContext *oc, const AVCodec *codec,
                       OutputStream *ost, AVDictionary *opt_arg)
{
    AVCodecContext *c;
    int nb_samples;
    int ret;
    AVDictionary *opt = NULL;

    c = ost->enc;

    /* open it */
    av_dict_copy(&opt, opt_arg, 0);
    ret = avcodec_open2(c, codec, &opt);
    av_dict_free(&opt);
    if (ret < 0) {
        fprintf(stderr, "Could not open audio codec: %s\n", ret);
        exit(1);
    }

    /* init signal generator */
    ost->t     = 0;
    ost->tincr = 2 * M_PI * 110.0 / c->sample_rate;
    /* increment frequency by 110 Hz per second */
    ost->tincr2 = 2 * M_PI * 110.0 / c->sample_rate / c->sample_rate;

    if (c->codec->capabilities & AV_CODEC_CAP_VARIABLE_FRAME_SIZE)
        nb_samples = 10000;
    else
        nb_samples = c->frame_size;

    ost->frame     = alloc_audio_frame(c->sample_fmt, c->channel_layout,
                                       c->sample_rate, nb_samples);
    ost->tmp_frame = alloc_audio_frame(AV_SAMPLE_FMT_S16, c->channel_layout,
                                       c->sample_rate, nb_samples);

    /* copy the stream parameters to the muxer */
    ret = avcodec_parameters_from_context(ost->st->codecpar, c);
    if (ret < 0) {
        fprintf(stderr, "Could not copy the stream parameters\n");
        exit(1);
    }

    /* create resampler context */
    ost->swr_ctx = swr_alloc();
    if (!ost->swr_ctx) {
        fprintf(stderr, "Could not allocate resampler context\n");
        exit(1);
    }

    /* set options */
    av_opt_set_int       (ost->swr_ctx, "in_channel_count",   c->channels,       0);
    av_opt_set_int       (ost->swr_ctx, "in_sample_rate",     c->sample_rate,    0);
    av_opt_set_sample_fmt(ost->swr_ctx, "in_sample_fmt",      AV_SAMPLE_FMT_S16, 0);
    av_opt_set_int       (ost->swr_ctx, "out_channel_count",  c->channels,       0);
    av_opt_set_int       (ost->swr_ctx, "out_sample_rate",    c->sample_rate,    0);
    av_opt_set_sample_fmt(ost->swr_ctx, "out_sample_fmt",     c->sample_fmt,     0);

    /* initialize the resampling context */
    if ((ret = swr_init(ost->swr_ctx)) < 0) {
        fprintf(stderr, "Failed to initialize the resampling context\n");
        exit(1);
    }
}


void video_encoder::encode_one_frame()
{
    if (enc_inf.encode_video || enc_inf.encode_audio) {
        /* select the stream to encode */
       if (enc_inf.encode_video &&
            (!enc_inf.encode_audio || av_compare_ts(enc_inf.video_st.next_pts, enc_inf.video_st.enc->time_base,
                                            enc_inf.audio_st.next_pts, enc_inf.audio_st.enc->time_base) <= 0)) {
            enc_inf.encode_video = !write_video_frame(enc_inf.oc, &enc_inf.video_st);
        } else {
            std::cout << "Encoding audio" << std::endl;
            enc_inf.encode_audio = !write_audio_frame(enc_inf.oc, &enc_inf.audio_st);
        }
    }
}

int video_encoder::write_audio_frame(AVFormatContext *oc, OutputStream *ost)
{
    AVCodecContext *c;
    AVFrame *frame;
    int ret;
    int dst_nb_samples;

    c = ost->enc;

    frame = audio_frame;//get_audio_frame(ost);

    if (frame) {
        /* convert samples from native format to destination codec format, using the resampler */
        /* compute destination number of samples */
        dst_nb_samples = av_rescale_rnd(swr_get_delay(ost->swr_ctx, c->sample_rate) + frame->nb_samples,
                                        c->sample_rate, c->sample_rate, AV_ROUND_UP);
        //av_assert0(dst_nb_samples == frame->nb_samples);

        /* when we pass a frame to the encoder, it may keep a reference to it
         * internally;
         * make sure we do not overwrite it here
         */
        ret = av_frame_make_writable(ost->frame);
        if (ret < 0)
            exit(1);

        /* convert to destination format */
        ret = swr_convert(ost->swr_ctx,
                          ost->frame->data, dst_nb_samples,
                          (const uint8_t **)frame->data, frame->nb_samples);
        if (ret < 0) {
            fprintf(stderr, "Error while converting\n");
            exit(1);
        }
        frame = ost->frame;

        frame->pts = av_rescale_q(ost->samples_count, (AVRational){1, c->sample_rate}, c->time_base);
        ost->samples_count += dst_nb_samples;
    }

    return write_frame(oc, c, ost->st, frame, ost->tmp_pkt);
}
void video_encoder::set_audio_frame(AVFrame* frame)
{
    audio_frame = frame;
}


  


  Normally the muxing example above uses get_audio_frame(ost) for frame insde write_audio_frame to create a dummy audio frame, but I want to use the audio that I have decoded from my input video. After decoding the audio frame I pass it to the encoder using set_audio_frame so my encoder can use it. Then I removed get_audio_frame(ost) and simply replaced it with audio_frame. So here is what my main loop looks like :

  


  ...
open_audio(args);
...
while(current_second < ouput_duration)
{
...
   video_reader_read_frame(buffer, &pts, start_ts);
   edit_decoded_video(buffer);
   ...
   if(frame_type == 2)
      encoder->set_audio_frame(audio_test->get_frame());
   encoder->encode_one_frame();
}


  


  And here is what my decoder looks like :

  


     int video_decode::decode_audio(AVCodecContext *dec, const AVPacket *pkt)
    {
        auto& frame= state.av_frame;
        int ret = 0;
    
        // submit the packet to the decoder
        ret = avcodec_send_packet(dec, pkt);
        if (ret < 0) {
            std::cout << "Error submitting a packet for decoding" << std::endl;
            return ret;
        }
    
        // get all the available frames from the decoder
        while (ret >= 0) {
            ret = avcodec_receive_frame(dec, frame);
            if (ret < 0) {
                // those two return values are special and mean there is no output
                // frame available, but there were no errors during decoding
                if (ret == AVERROR_EOF || ret == AVERROR(EAGAIN))
                    return 0;
    
                std::cout << "Decode err" << std::endl;
                return ret;
            }
    
            if (ret < 0)
                return ret;
        }
    
        return 0;
    }

    int video_decode::video_reader_read_frame(uint8_t* frame_buffer, int64_t* pts, double seg_start) 
    {
        // Unpack members of state
        auto& width = state.width;
        auto& height = state.height;
        auto& av_format_ctx = state.av_format_ctx;
        auto& av_codec_ctx = state.av_codec_ctx;
        auto& audio_codec_ctx = state.audio_codec_ctx;
        auto& video_stream_index = state.video_stream_index;
        auto& audio_stream_index = state.audio_stream_index;
        auto& av_frame = state.av_frame;
        auto& av_packet = state.av_packet;
        auto& sws_scaler_ctx = state.sws_scaler_ctx;

        // Decode one frame
        //double pt_in_seconds = (*pts) * (double)state.time_base.num / (double)state.time_base.den;
        if (!this->skipped) {
            this->skipped = true;
            *pts = (int64_t)(seg_start * (double)state.time_base.den / (double)state.time_base.num);
            video_reader_seek_frame(*pts);
        }

        int response;
        while (av_read_frame(av_format_ctx, av_packet) >= 0) {
            // Audio decode

            if (av_packet->stream_index == video_stream_index){
                std::cout << "Decoded VIDEO" << std::endl;

                response = avcodec_send_packet(av_codec_ctx, av_packet);
                if (response < 0) {
                    printf("Failed to decode packet: %s\n", av_make_error(response));
                    return false;
                }


                response = avcodec_receive_frame(av_codec_ctx, av_frame);
                if (response == AVERROR(EAGAIN) || response == AVERROR_EOF) {
                    av_packet_unref(av_packet);
                    continue;
                } else if (response < 0) {
                    printf("Failed to decode packet: %s\n", av_make_error(response));
                    return false;
                }


                *pts = av_frame->pts;
                // Set up sws scaler
                if (!sws_scaler_ctx) {
                    auto source_pix_fmt = correct_for_deprecated_pixel_format(av_codec_ctx->pix_fmt);
                    sws_scaler_ctx = sws_getContext(width, height, source_pix_fmt,
                                                    width, height, AV_PIX_FMT_RGB0,
                                                    SWS_BICUBIC, NULL, NULL, NULL);
                }
                if (!sws_scaler_ctx) {
                    printf("Couldn't initialize sw scaler\n");
                    return false;
                }

                uint8_t* dest[4] = { frame_buffer, NULL, NULL, NULL };
                int dest_linesize[4] = { width * 4, 0, 0, 0 };
                sws_scale(sws_scaler_ctx, av_frame->data, av_frame->linesize, 0, height, dest, dest_linesize);
                av_packet_unref(av_packet);
                return 1;
            }
            if (av_packet->stream_index == audio_stream_index){
                std::cout << "Decoded AUDIO" << std::endl;
                decode_audio(audio_codec_ctx, av_packet);
                av_packet_unref(av_packet);
                return 2;
            }else {
                av_packet_unref(av_packet);
                continue;
            }

            av_packet_unref(av_packet);
            break;
        }


        return true;
    }
void init()
{
...
if (open_codec_context(&audio_stream_index, &audio_dec_ctx, av_format_ctx, AVMEDIA_TYPE_AUDIO) >= 0) {
      audio_stream = av_format_ctx->streams[audio_stream_index];
}
...
}


  


  My decoder uses the same format context, packet and frame for video and audio decoding and uses separate stream and codec context.

  


  Why am I getting Queue input is backward in time error and how can I propperly encode the audio ? I'm not sure but from the looks of it decodes the audio just fine. And again there are no problems on video encoding/decoding whatsoever.

  


• aacenc_pred : rework the way prediction is done

  29 août 2015, par Rostislav Pehlivanov

  aacenc_pred : rework the way prediction is done
  This commit completely alters the algorithm of prediction.
  
  The original commit which introduced prediction was completely
  
  incorrect to even remotely care about what the actual coefficients
  
  contain or whether any options were enabled. Not my actual fault.
  This commit treats prediction the way the decoder does and expects
  
  to do : like lossy encryption. Everything related to prediction now
  
  happens at the very end but just before quantization and encoding
  
  of coefficients. On the decoder side, prediction happens before
  
  anything has had a chance to even access the coefficients.
  Also the original implementation had problems because it actually
  
  touched the band_type of special bands which already had their
  
  scalefactor indices marked and it’s a wonder the asserion wasn’t
  
  triggered when transmitting those.
  Overall, this now drastically increases audio quality and you should
  
  think about enabling it if you don’t plan on playing anything encoded
  
  on really old low power ultra-embedded devices since they might not
  
  support decoding of prediction or AAC-Main. Though the specifications
  
  were written ages ago and as times change so do the FLOPS.
  Signed-off-by : Rostislav Pehlivanov <atomnuker@gmail.com>
  

  • [DH] libavcodec/aac.h
  • [DH] libavcodec/aaccoder.c
  • [DH] libavcodec/aacenc.c
  • [DH] libavcodec/aacenc.h
  • [DH] libavcodec/aacenc_pred.c
  • [DH] libavcodec/aacenc_pred.h

• Very long ffmpeg command due to drawtext commands for every second

  24 avril 2022, par principal-ideal-domain

  I have a video of about 20 minutes length and want to show every second a different text using the drawtext filter. I used a java software to compute a very long ffmpeg command (more than 100,000 characters long). Pasting it into the PowerShell took a long time and then I got the error

  


  Program 'ffmpeg.exe' failed to run: The filename or extension is too longAt line:1 char:1


  


  So the command is obviously too long. Can I somehow outsource it into an external file ? I'm not looking for the subtitles filter instead of the drawtext filter because I'm using special functionalities of drawtext.