Recherche avancée

Sur d’autres sites (7462)

• How to solve error : "cvCreateFileCaptureWithPreference : backend FFMPEG doesn't support legacy API anymore"

  24 octobre 2018, par Sarvin

  I am trying to build and run an OpenCV project from Github using OpenCV 3.4.2, Cmake 3.13.0-rc1 and VS 2017. The project builds successfully as I understand from the build output but it throws the above warning after I run the program. The video I want to load is in a .AVI format and uses cvCaptureFromAVI function. Just trying to learn OpenCV, your help and kindness is appreciated.

  Source :

  #include <iostream>
  
  #include <opencv2></opencv2>opencv.hpp>
  
  
  
  #include "package_bgs/PBAS/PixelBasedAdaptiveSegmenter.h"
  
  #include "package_tracking/BlobTracking.h"
  
  #include "package_analysis/VehicleCouting.h"
  
  
  
  using namespace cv;
  
  
  
  int main(int argc, char **argv)
  
  {
  
    std::cout &lt;&lt; "Using OpenCV " &lt;&lt; CV_MAJOR_VERSION &lt;&lt; "." &lt;&lt; 
  
    CV_MINOR_VERSION &lt;&lt; "." &lt;&lt; CV_SUBMINOR_VERSION &lt;&lt; std::endl;
  
  
  
    /* Open video file */
  
    CvCapture *capture = 0;
  
  
  
    capture = cvCaptureFromAVI("video.avi");
  
    if(!capture){
  
      std::cerr &lt;&lt; "Cannot open video ting tong!" &lt;&lt; std::endl;
  
      return 1;
  
    }
  
  
  
    /* Background Subtraction Algorithm */
  
    IBGS *bgs;
  
    bgs = new PixelBasedAdaptiveSegmenter;
  
  
  
    /* Blob Tracking Algorithm */
  
  
  
  
  
  
  
  cv::Mat img_blob;
  
    BlobTracking* blobTracking;
  
    blobTracking = new BlobTracking;
  
  
  
    /* Vehicle Counting Algorithm */
  
    VehicleCouting* vehicleCouting;
  
    vehicleCouting = new VehicleCouting;
  
  
  
    std::cout &lt;&lt; "Press 'q' to quit..." &lt;&lt; std::endl;
  
    int key = 0;
  
    IplImage *frame;
  
    while(key != 'q')
  
    {
  
      frame = cvQueryFrame(capture);
  
      if(!frame) break;
  
  
  
      cv::Mat img_input = cv::cvarrToMat(frame);
  
      cv::imshow("Input", img_input);
  
  
  
      // bgs->process(...) internally process and show the foreground mask image
  
      cv::Mat img_mask;
  
      bgs->process(img_input, img_mask);
  
  
  
      if(!img_mask.empty())
  
      {
  
        // Perform blob tracking
  
        blobTracking->process(img_input, img_mask, img_blob);
  
  
  
        // Perform vehicle counting
  
        vehicleCouting->setInput(img_blob);
  
        vehicleCouting->setTracks(blobTracking->getTracks());
  
        vehicleCouting->process();
  
      }
  
  
  
      key = cvWaitKey(1);
  
    }
  
  
  
    delete vehicleCouting;
  
    delete blobTracking;
  
    delete bgs;
  
  
  
    cvDestroyAllWindows();
  
    cvReleaseCapture(&amp;capture);
  
  
  
    return 0;
  
  }
  
  </iostream>

  Error :

  

• Why do I get "error : invalid instruction mnemonic" when compiling FFmpeg for Android ?

  20 octobre 2018, par donturner

  I’m trying to compile FFmpeg 4.0 for Android using clang from the NDK (r18.1.5063045). My host architecture is Mac OS.

  Here’s my configure command :

  ./configure \
  
  --target-os=android \
  
  --arch=armv7-a \
  
  --enable-cross-compile \
  
  --cc=${ANDROID_NDK}/toolchains/llvm/prebuilt/darwin-x86_64/bin/clang \
  
  --enable-shared \
  
  --disable-static 

  This completes successfully. So I run make -j8 and get the following error :

  In file included from libavfilter/vf_mestimate.c:22:
  
  In file included from ./libavcodec/mathops.h:40:
  
  ./libavcodec/arm/mathops.h:90:26: error: invalid instruction mnemonic 'itt'
  
      "cmp   %1, %2  \n\t"
  
                       ^
  
  <inline asm="asm">:3:2: note: instantiated into assembly here
  
      itt   gt      
  
      ^~~
  
  </inline>

  Any idea what could be causing this ?

• FFmpeg - filter_complex error,Too many inputs specified for the "scale" filter

  20 septembre 2018, par wensefu

  I’m trying to concat some video into one single video using ffmpeg. here’s my ffmpeg cmd :

  ffmpeg -i input1.avi -i input2.mkv -filter_complex [0:v]scale=1920:1080:force_original_aspect_ratio=decrease,pad=1920:1080:(ow-iw)/2:(oh-ih)/2[v0];[v0][0:a][1:v]scale=1920:1080:force_original_aspect_ratio=decrease,pad=1920:1080:(ow-iw)/2:(oh-ih)/2[v1];[v1][1:a]concat=n=2:v=1:a=1[outv][outa] -map [outv] -map [outa] -vcodec libx264 -crf 27 -preset ultrafast out.mp4 

  but got error :

  [AVFilterGraph @ 0xd177f500] Too many inputs specified for the "scale" filter.
  
  E AVLOG   : Error initializing complex filters.
  
  E AVLOG   : Invalid argument
  
  E AVLOG   : At least one output file must be specified
  
  I AVLOG   : Stream mapping:
  
  I AVLOG   : Press [q] to stop, [?] for help
  
  I AVLOG   : No more output streams to write to, finishing.
  
  E AVLOG   : 
  
  F libc    : Fatal signal 11 (SIGSEGV), code 1, fault addr 0x0 in tid 17256 (IntentService[v)

  Any ideas anyone ?

  Many thanks in advance.