Recherche avancée

Sur d’autres sites (9266)

• Saving video using OpenCV

  17 avril 2014, par Prakhar Mohan Srivastava

  Given below is the Python script I am using for accessing an IP Camera and saving the incoming video stream :

  import cv2
  
  from cv2 import cv
  
  writer=cv2.VideoWriter("vid.avi",cv.CV_FOURCC('X', '2', '6', '4'), 24, (320,240))
  
  def cvloop():    
  
      stream=urllib.urlopen('URL of the camera')
  
      bytes=''
  
      while True:
  
          bytes+=stream.read(1024)
  
          a = bytes.find('\xff\xd8')
  
          b = bytes.find('\xff\xd9')
  
          if a!=-1 and b!=-1:
  
              jpg = bytes[a:b+2]
  
              bytes= bytes[b+2:]
  
              i = cv2.imdecode(np.fromstring(jpg, dtype=np.uint8),cv2.CV_LOAD_IMAGE_COLOR)            
  
              tki = ImageTk.PhotoImage(Image.fromarray(cv2.cvtColor(i, cv2.COLOR_BGR2RGB)))
  
              cv2.imshow('i',i)
  
              writer.write(i)
  
              if cv2.waitKey(1) ==27:
  
                  exit(0)

  Now, if you want to know why I am parsing the jpg images from the stream the answer can be found here. The answer explains everything.

  As for the video writer script, I have referred OpenCV's documentation which can be found here.

  Now instead of using the default list of codecs I am using H-264 but I have tried all the codecs installed on my system.

  Question : Now the problem is on executing the program the video file is being made, on running it I see nothing. Also, on using the software GSpot to analyse the video, I am getting the number of frames in the video. Same is the case with FFMPEG. Where am I going wrong ?

  Camera: Axis IP Camera M1103
  
  OS: Windows 7 32bit

• fate : add unit test for dnn-layer-pad

  29 juillet 2019, par Guo, Yejun

  fate : add unit test for dnn-layer-pad
  'make fate-dnn-layer-pad' to run the test
  Signed-off-by : Guo, Yejun <yejun.guo@intel.com>
  
  Signed-off-by : Pedro Arthur <bygrandao@gmail.com>
  

  • [DH] tests/Makefile
  • [DH] tests/dnn/Makefile
  • [DH] tests/dnn/dnn-layer-pad-test.c
  • [DH] tests/fate/dnn.mak

• libavfilter/dnn : support multiple outputs for tensorflow model

  25 avril 2019, par Guo, Yejun

  libavfilter/dnn : support multiple outputs for tensorflow model
  some models such as ssd, yolo have more than one output.
  the clean up code in this patch is a little complex, it is because
  
  that set_input_output_tf could be called for many times together
  
  with ff_dnn_execute_model_tf, we have to clean resources for the
  
  case that the two interfaces are called interleaved.
  Signed-off-by : Guo, Yejun <yejun.guo@intel.com>
  
  Signed-off-by : Pedro Arthur <bygrandao@gmail.com>
  

  • [DH] libavfilter/dnn_backend_native.c
  • [DH] libavfilter/dnn_backend_native.h
  • [DH] libavfilter/dnn_backend_tf.c
  • [DH] libavfilter/dnn_backend_tf.h
  • [DH] libavfilter/dnn_interface.h
  • [DH] libavfilter/vf_sr.c