Recherche avancée

Sur d’autres sites (15790)

• How to send continuous stream of frames to server most efficiently

  27 avril 2019, par Duthopi

  I am trying to send frames from a local camera (raspberry pi camera, but could also be my laptop’s webcam) to a Google cloud instance, on which I am running AI processing of the frames.

  I am managing to send frames captured through opencv via http (i.e. tcp ??) and receiving them on a flask server. When the flask server is running locally I can get good fps (50+ fps for image size 640x480), however once I send the frames to a flask app on the google instance the fps drop drastically to 5fps.

  How I currently send frames :

  while True:
  
          frame = vs.read() #Separate thread, using cv2 to get the frame
  
  
  
          ret, jpeg = cv2.imencode('.jpg', frame)
  
          imgdata = jpeg.tobytes()
  
          response = requests.post(
  
              url='http://<ip address="address" of="of" google="google" instance="instance">:<port>',
  
              data= imgdata,
  
              headers={'content-type':'image/jpeg'},
  
              )
  
  </port></ip>

  I see two problems with this :
  1 - using tcp means I am slower than udp protocol, however udp is limited in byte size. Correct me if I am wrong, but it seems very complex to send truncated frames and put them back together on the server..
  2 - Even if I had udp working, there is no compression of frames, so I will never reach an efficient transfer

  I expect the answer to be something like using ffmpeg, but so far I only figured out how to stream frames on a local port with ffmpeg, I do not know if it is possible to send frames to a remote server.

  Any recommendations on the best way forward ?

• Is it possible to pipe an ffmpeg output with multiple files (HLS or DASH)

  29 août 2020, par New Dev

  I'm using FFmpeg to generate fragmented MP4s in a dash and HLS format.

  


  ffmpeg -i input.mov -f dash -seg_duration 6 -hls_playlist true output.mbd


  


  The above (simplified) command outputs multiple files, in addition to output.mdb (e.g. actual segments, master.m3u8, etc...)

  


  Is there a way to get each produced file into their individual and separate output streams ?

  


  ---

  


  Broader context :

  


  I'm trying to build a transcoder in Node.js running in Google Cloud, with the idea being that it writes directly to a Google Storage through a writable stream. I can only create a stream per file, but since the number of files is dynamic, I'm not sure how to obtain a stream from each file.

  


• src/libFLAC/stream_decoder.c : Fix buffer read overflow.

  18 novembre 2014, par Erik de Castro Lopo

  src/libFLAC/stream_decoder.c : Fix buffer read overflow.
  This is CVE-2014-8962.
  Reported-by : Michele Spagnuolo,
  
               Google Security Team <mikispag@google.com>
  

  • [DH] src/libFLAC/stream_decoder.c