Recherche avancée

Sur d’autres sites (11186)

• Cannot get MediaStream from Electron app to browser via Openvidu media server

  11 juin 2021, par foxy.bunny

  I'm currently having an issue with openvidu-browser-2.17.0.js

  


  I'm trying to transfer a camera stream from RTSP to an Openvidu media server deployed on cloud and get the stream back on the browser.

  


  The RTSP stream was converted into an HLS stream using FFmpeg and played using hls.js. It was captured in an video tag in HTML and I used the HTMLMediaElement.captureStream().getVideoTrack() to generate MediaStreamTrack and passed it into the videoSource property in initPublisher. This publisher part was wrapped in electron. Then, the stream was connected to our cloud Openvidu server deployed on-premises exactly like in the Docs (https://docs.openvidu.io/en/2.18.0/deployment/deploying-on-premises/). The subscriber part is a simple HTML page displayed on the browser to get the stream from the cloud server. The subscriber is responsible for creating session and generating stream from the media server when streamCreated event occurs.

  


  The whole workflow worked nicely when we tested it with our webcam, however, when we use MediaStreamTrack of our stream video instead of webcam, the subscriber part only showed the blank video.

  


  My question is :

  


  1 : Is it possible to stream an MediaStream to Openvidu media server like that ?

  


  2 : If yes, then what am I doing wrong here ?

  


  Describe the bug

  


  Cannot get MediaStream from Electron app to browser via Openvidu media server.

  


  Expected behavior

  


  Receive the stream from the Electron app to browser via Openvidu media server.

  


  Wrong current behavior

  


  Only get blank video

  


  Client device info

  


  

  • Chrome Version 91.0.4472.77 (Official Build) (64-bit) on Windows 10 Version 10.0.19042 Build 19042

  


  


• Right way to use vmaf with ffmpeg

  14 juin, par dravit

  I am trying to calculate the VMAF score of a processed video wrt the original file.

  


  Command I have used :

  


  ffmpeg -y -loglevel info -stats -i original.mp4 -i processed.mp4 -lavfi "[0]null[refdeint];[refdeint]scale=1920:1080:flags=neighbor[ref];[1]setpts=PTS+0.0/TB[b];[b]scale=1920:1080:flags=neighbor[c];[c][ref]libvmaf=log_fmt=json:phone_model=1:model_path={model_path_here}/vmaf_v0.6.1.json:n_subsample=1:log_path=log.json" -f null -


  


  Now as per the official documentation of vmaf with ffmpeg found here, it says source/reference file followed by the encoded/distorted/processed file.

  


  But almost all of the blogs I came across, they are using the other way round order of the args, i.e. processed file followed by the original file.

  


  Few examples :

  


  


  1. https://medium.com/@eyevinntechnology/keep-an-eye-on-the-video-quality-b9bcb58dd5a1 : search for "Using VMAF within FFMPEG" in it.

     


  


  2. https://websites.fraunhofer.de/video-dev/calculating-vmaf-and-psnr-with-ffmpeg/ : search for "Metric Calculation with FFmpeg" in it.

     


  


  


  EDIT

  


  NOTE : Changing the order does change the VMAF score.

  


• FFmpeg decode OGG Vorbis audio as float, how to get float data from decoder

  19 novembre 2020, par cs guy

  I am trying to decode a OGG Vorbis audio file with the following :

  


  // if opening input failed
    if (avformat_open_input(&pFormatContext, filePath.c_str(), av_find_input_format("ogg"),
                            nullptr) != 0) {
        
        LOGE("FFmpegExtractor can not open the file! %s", filePath.c_str());
        return FAILED_TO_LOAD;
    }

    durationInMillis = pFormatContext->duration / AV_TIME_BASE * secondsToMilli;
    LOGD("FFmpegExtractor opened the file, Duration: %llu", durationInMillis);

    avformat_find_stream_info(pFormatContext, nullptr);

    LOGD("pFormatContext streams num: %ui", pFormatContext->nb_streams);

    int sendPacketResult = 0;
    int receiveFrameResult = 0;

    for (int i = 0; i < pFormatContext->nb_streams; i++) {
        AVCodecParameters *pLocalCodecParameters = pFormatContext->streams[i]->codecpar;

        if (pLocalCodecParameters->codec_type == AVMEDIA_TYPE_AUDIO) {
            avCodecContext = avcodec_alloc_context3(avCodec);
            if (!avCodecContext) {
                
                LOGE("FFmpegExtractor avcodec_alloc_context3 failed!");
                return FAILED_TO_LOAD;
            }

            if (avcodec_parameters_to_context(avCodecContext, pLocalCodecParameters) < 0) {
                
                LOGE("FFmpegExtractor avcodec_parameters_to_context failed!");
                return FAILED_TO_LOAD;
            }

            if (avcodec_open2(avCodecContext, avCodec, nullptr) < 0) {
                
                LOGE("FFmpegExtractor avcodec_open2 failed!");
                return FAILED_TO_LOAD;
            }
        }

        while (av_read_frame(pFormatContext, avPacket) >= 0) {
            
            sendPacketResult = avcodec_send_packet(avCodecContext, avPacket);

            
            receiveFrameResult = avcodec_receive_frame(avCodecContext, avFrame);

            // TODO: Get frames PCM data from the decoder, 
        }
    }


  


  I am stuck on the last part. I want to get the decoded data of the current frame as a float between -1 to +1 for all the channels the audio has but I have no idea how to. I looked at the official documents but I could not understand them. How can I get the data from the decoder on the line

  


  


  // TODO : Get frames PCM data from the decoder,