Recherche avancée

Sur d’autres sites (8190)

• BGRA to YUV420 FFmpeg giving bad output

  6 octobre 2020, par NewbieCoder

  My purpose is a screen recorder. I am using the Windows DXGI API to receive screenshots and I'm encoding the screenshots into a video using libx264. Feeding BGRA images directly to libx264 is producing weird colors in the output video. So, to get correct colors, I am trying to convert the BGRA to YUV420p. To speed up encoding, I am also trying to downscale the BGRA image.

  


  So I am getting an 1920x1080 BGRA image and I want to convert it to 1280x720 YUV420p. For this, I am using FFmpeg swscale library to do both the format conversion and downscaling.

  


  The problem is that the output video is coming like 3 images in the same frame. Please see this video. https://imgur.com/a/EYimjrJ

  


  I tried just BGRA to YUV conversion without any downscaling and it is working fine. But BGRA to YUV with downscaling is giving this problem.

  


  What is the cause for this problem ? How do I fix it ?

  


  Here is my code snippet :

  


  uint8_t* Image;
x264_picture_t picIn, picOut;
x264_picture_alloc(&picIn, X264_CSP_I420, 1280, 720);

SwsContext* sws = sws_getContext(1920, 1080, AV_PIX_FMT_BGRA, 1280, 720, AV_PIX_FMT_YUV420P, SWS_BILINEAR, NULL, NULL, NULL);

while (true)
{
    take_screenshot(&Image);

    AVFrame BGRA;
    BGRA.linesize[0] = 1280 * 4;
    BGRA.data[0] = Image;
    sws_scale(sws, BGRA.data, BGRA.linesize, 0, 1080, picIn.img.plane, picIn.img.i_stride);

    nal_size = x264_encoder_encode(h, &nals, &nal_count, &picIn, &picOut);
    save_to_flv(nals, nal_size, nal_count);
}


  


  Here are my libx264 parameters :

  


          x264_param_default_preset(&param, preset, 0);
        param.i_csp = X264_CSP_I420;
        param.i_width = 1280;
        param.i_height = 720;
        param.i_fps_num = 30;
        param.i_fps_den = 1;
        param.rc.i_bitrate = 2500;
        param.i_bframe = 0;
        param.b_repeat_headers = 0;
        param.b_annexb = 1;     

        x264_param_apply_profile(&param, 0);
        h = x264_encoder_open(&param);


  


• libx264 codex not enabled in ffmpeg ubuntu 18.04

  23 novembre 2022, par Bendemann

  I have some weird codex installation issues with the following docker image.

  


  ARG PYTORCH="1.8.0"
ARG CUDA="11.1"
ARG CUDNN="8"
FROM pytorch/pytorch:${PYTORCH}-cuda${CUDA}-cudnn${CUDNN}-devel

ENV TORCH_CUDA_ARCH_LIST="6.0 6.1 7.0 8.0+PTX"
ENV TORCH_NVCC_FLAGS="-Xfatbin -compress-all"
ENV CMAKE_PREFIX_PATH="$(dirname $(which conda))/../"

# https://github.com/NVIDIA/nvidia-docker/issues/1632
RUN apt-key del 7fa2af80
RUN apt-key adv --fetch-keys https://developer.download.nvidia.com/compute/cuda/repos/ubuntu2004/x86_64/3bf863cc.pub
RUN apt-key adv --fetch-keys https://developer.download.nvidia.com/compute/machine-learning/repos/ubuntu2004/x86_64/7fa2af80.pub

RUN apt-get update && apt-get install -y \
        git nano ninja-build p7zip-full imagemagick wget unzip \
        libglib2.0-0 libsm6 libxrender-dev libxext6 libturbojpeg \
        libxrender1 libfontconfig1 freeglut3-dev llvm-6.0-tools curl \
        amqp-tools ffmpeg libx264-dev \
    && apt --fix-broken install \
    && apt-get clean \
    && rm -rf /var/lib/apt/lists/* \
    # for visualizing
    && wget https://github.com/mmatl/travis_debs/raw/master/xenial/mesa_18.3.3-0.deb \
    && dpkg -i ./mesa_18.3.3-0.deb || true \
    && apt install -f \
    && git clone https://github.com/mmatl/pyopengl.git \
    && pip install ./pyopengl


  


  First of all, libx264 is supposed to be installed by a simple apt-get install ffmpeg in ubuntu 18.04.5. Indeed I see that it is being installed in the installation instructions but for some reason, it's not enabled. This is confirmed when running ffmpeg -codecs | grep 264, which doesn't show libx264 (only h264, libopenh264 are there).

  


  In addition, I also compiled from source, explicitly enabling libx264 during installation. It didn't make a difference.

  


• libx264 codex not enabled in ffmpeg ubuntu 18.04

  23 novembre 2022, par Bendemann

  I have some weird codex installation issues with the following docker image.

  


  ARG PYTORCH="1.8.0"
ARG CUDA="11.1"
ARG CUDNN="8"
FROM pytorch/pytorch:${PYTORCH}-cuda${CUDA}-cudnn${CUDNN}-devel

ENV TORCH_CUDA_ARCH_LIST="6.0 6.1 7.0 8.0+PTX"
ENV TORCH_NVCC_FLAGS="-Xfatbin -compress-all"
ENV CMAKE_PREFIX_PATH="$(dirname $(which conda))/../"

# https://github.com/NVIDIA/nvidia-docker/issues/1632
RUN apt-key del 7fa2af80
RUN apt-key adv --fetch-keys https://developer.download.nvidia.com/compute/cuda/repos/ubuntu2004/x86_64/3bf863cc.pub
RUN apt-key adv --fetch-keys https://developer.download.nvidia.com/compute/machine-learning/repos/ubuntu2004/x86_64/7fa2af80.pub

RUN apt-get update && apt-get install -y \
        git nano ninja-build p7zip-full imagemagick wget unzip \
        libglib2.0-0 libsm6 libxrender-dev libxext6 libturbojpeg \
        libxrender1 libfontconfig1 freeglut3-dev llvm-6.0-tools curl \
        amqp-tools ffmpeg libx264-dev \
    && apt --fix-broken install \
    && apt-get clean \
    && rm -rf /var/lib/apt/lists/* \
    # for visualizing
    && wget https://github.com/mmatl/travis_debs/raw/master/xenial/mesa_18.3.3-0.deb \
    && dpkg -i ./mesa_18.3.3-0.deb || true \
    && apt install -f \
    && git clone https://github.com/mmatl/pyopengl.git \
    && pip install ./pyopengl


  


  First of all, libx264 is supposed to be installed by a simple apt-get install ffmpeg in ubuntu 18.04.5. Indeed I see that it is being installed in the installation instructions but for some reason, it's not enabled. This is confirmed when running ffmpeg -codecs | grep 264, which doesn't show libx264 (only h264, libopenh264 are there).

  


  In addition, I also compiled from source, explicitly enabling libx264 during installation. It didn't make a difference.