Recherche avancée

Sur d’autres sites (6753)

• opencv3.1 python VideoCapture black screen

  3 mai 2017, par alex

  i’m trying to use opencv3.1 with python3.5
  following this official)tutorial
  http://docs.opencv.org/3.0-beta/doc/py_tutorials/py_gui/py_video_display/py_video_display.html#display-video

  the camera led is "on" but the window doesn’t show any image (black)

  my current enviroment :
   windows 10
   python 3.5.2 (32bit)
   numpy 1.12.0b1 (32bit) binaries downloaded from : http://www.lfd.uci.edu/ gohlke/pythonlibs/
   opencv 3.1.0 (32bit) binaries downloaded from : http://www.lfd.uci.edu/ gohlke/pythonlibs/

  they seem installed correctly !

  as the tutorial says :

  Note Make sure proper versions of ffmpeg or gstreamer is installed. Sometimes, it is a headache to work with Video Capture mostly due to wrong installation of ffmpeg/gstreamer.

  i’ve no ffmpeg installed !!! so i suppose this is the cause of my problem
  so the questions are :
  1) which version of ffmpeg is needed ?
  2) how install ffmpeg on windows ? it seems a setup doesn’t exist(there is a binary section with a zip file ? how to use it after unzipped ?)
  3) is possible to see(in some way) if the opencv binaries(that i’ve downloaded) were compiled with ffmpeg support(flag ?)

  PS : i’ve tried to add the ffmpeg folder path the the PATH enviroment variable but nothing change

• python3 openCV VideoCapture only black image (Raspbian Stretch)

  21 juin 2018, par PrimuS

  I have a little python function that analyzes a video, chunks it into one image per second and gives me the most dominant color for that image. (Code can be found here : https://github.com/primus852/python-movie-barcode)

  This works great on my Windows testing environment. However, on my Rasbian Stretch Raspberry Pi Setup it only produces a black image, as the source seems to be black.

  I compiled OpenCV (3.4.1) myself with this great article : https://www.pyimagesearch.com/2017/09/04/raspbian-stretch-install-opencv-3-python-on-your-raspberry-pi/, and it worked perfectly fine. I am using python3 and a virtualenv.

  I tried adding the ffmpeg package : apt install ffmpeg, to no avail.

  2 ideas

  • I compiled the openCV source without support for mkv/mp4/similar ? If so, how would I check that and can I just "re-compile" the whole package ?
  • I am missing codecs, where would I be able to check that ?

  The crucial code (I think) is this :

  cap = cv2.VideoCapture(full_path) are there other options that do not break the majority of my code ? I read about skvideo.io from scikit-video but that does not seem to work with my code...

  I am new to python, any hint is appreciated

  // EDIT, I think it is not a duplicate because I pass the I get no error that the capture cannot be opened and :

  OpenCV FFMPEG support :

  • python -c "import cv2; print(cv2.getBuildInformation())" | grep -i ffmpeg
    • returns : FFMPEG: YES

  FFMPEG Codec :

  • ffmpeg -codecs | grep -i avc (file is using AVC)
    • returns : DEV.LS h264 H.264 / AVC / MPEG-4 AVC / MPEG-4 part 10 (decoders: h264 h264_mmal h264_vdpau ) (encoders: libx264 libx264rgb h264_omx h264_vaapi )

  File PATH

  It exists and the path is correct...

  Any other ideas ? Could it possibly the virtualenv ?

  //EDIT2

  Tried with an AVI file

  works...

• Black video when using ffplay of received data over upd [closed]

  4 février 2023, par Alexandru-Marian Buza

  I'm encoding frames from a id3d11texture2d, and the receive packet size is around 70-80.
When i'm sending over udp, and use ffplay to show the video, i get a black screen.
Resolution and fps are well determined by ffplay.

  


  AcquiredBuffer.Attach(Buffer.MetaData.pSurface);

            ID3D11Texture2D* texture;
            ID3D11Texture2D* stagingTexture;
            AcquiredBuffer->QueryInterface(IID_PPV_ARGS(&texture));
            
            texture->GetDesc(&desc);
            desc.Usage = D3D11_USAGE_STAGING;
            desc.CPUAccessFlags = D3D11_CPU_ACCESS_READ;
            desc.Format = DXGI_FORMAT_R8G8B8A8_UNORM;
            desc.BindFlags = 0;
            desc.MiscFlags = 0;
            hr = m_Device->Device->CreateTexture2D(&desc, NULL, &stagingTexture);
            if (stagingTexture != 0) {
                m_Device->DeviceContext->CopyResource(stagingTexture, texture);
                D3D11_MAPPED_SUBRESOURCE mappedData;
                m_Device->DeviceContext->Map(stagingTexture, 0, D3D11_MAP_READ, 0, &mappedData);
                
                memcpy(frame->data[0], mappedData.pData, mappedData.DepthPitch);
                auto ret = avcodec_send_frame(codec, frame);
                if (ret < 0) {
                    writeText("Send frame failed");
                }
                else {
                    while (ret >= 0) {
                        ret = avcodec_receive_packet(codec, packet);
                        for (int i = 0; i < packet->size; i++) {
                            if (packet->data[i] != 0) {
                            }
                        }
                        if (ret == AVERROR(EAGAIN) || ret == AVERROR_EOF) {
                            break;
                        }
                        else if (ret < 0) {
                            writeText("Error durring encoding");
                        }
                        if (sendto(s, (char*)packet->data, packet->size * sizeof(uint8_t), 0, (sockaddr*)&dest, sizeof(dest)) == SOCKET_ERROR) {
                            writeText("Failed to send frame");
                        }
                        av_packet_unref(packet);
                    }
                }
                
                m_Device->DeviceContext->Unmap(stagingTexture, 0);
                stagingTexture->Release();