Recherche avancée

Sur d’autres sites (6599)

• Using ffmpeg rtmp stream a static image and audio input [on hold]

  30 octobre 2017, par Chad

  Using a Raspberry Pi, stream audio in and use a static image as the video input thru ffmpeg over RTMP to a Cloud video provider (DaCast in this instance)

  So far, I’ve gone through many blog posts, Stack Overflow questions, and package documentation. I’ve found that most of the posts are no longer valid with the newer versions of ffmpeg. Or don’t quite line up with what I am trying to achieve.

  However, I have figured out the right settings to stream the Raspberry Pi Camera v2 with the audio in.

  ffmpeg \
    -f alsa -ac 1 -i plughw:1,0 \
    -f v4l2 -s 1920x1080 -r 30 -input_format h264 -i /dev/video0 \
    -vcodec copy -preset veryfast -r 15 -g 30 -b:v 64k -ar 44100 -threads 6 -b:a 96k -bufsize 3000k \
    -f flv rtmp ://streaming_server_url
  

  But can’t seem to get it right to replace the video input with a static image.

  I have tried removing the 3rd line and adding -loop 1 -i '/path/to/image.jpg'

  The logs look like :

  [alsa @ 0x55e4b980] Thread message queue blocking ; consider raising the thread_queue_size option (current value : 1024)
  [alsa @ 0x55e4b980] ALSA buffer xrun.  0kB time=00:00:00.00 bitrate=N/A speed=   0x
  [alsa @ 0x55e4b980] ALSA buffer xrun.130kB time=00:00:00.27 bitrate=3822.5kbits/s speed=0.0403x
  ...
  

  I have also tried looping a 4 second video, with similar outcomes.

  ---

  My Setup for context :

  • Raspberry Pi 3 Model B
  • USB Audio Device (Sabrent USB External Stereo Sound Adapter)
  • Ubuntu MATE 16.04.2 (Xenial)
  • ffmpeg version 3.2-2+rpi1 xenial1.7 (I can post what is configured with the build, if needed)

• Trying to grab video stream from a 802W device

  1er juin 2015, par brentil

  A group of us in the RC hobby forums had started trying to use a device called the 802W, it takes RCA in and then broadcasts it back out over a WiFi you connect to via an Android or iOS device. They’re typically used for backup camera addon systems for vehicles. We want to use it to do FPV (First Person Video/View) with using smartphones instead of buying more expensive FPV goggles.

  802W device example (plenty of clones online)

  http://www.amazon.com/Wireless-Backup-Camera-Transmitter-Android/dp/B00LJPTJSY

  The problem is you can only use their application WIFI_AVIN or WIFI_AVIN2 from the app stores to connect to it because they don’t publish the information about how to grab the stream data. We want to write our own apps that can use the stream to better show the information. We’ve tried using VLC to grab the stream from an Android phone or a Windows PC but we’ve had no success so far. I was hoping someone could look at the Wireshark outputs and might understand what they’re looking at better than I am. I "think" it’s a UDP multicast being broadcasted but I just don’t know enough to be sure. We’ve tried using VLC to connect to network streams directly on the device or from udp ://@ type addresses but I think part of the issue too might be we’re missing the file path of the stream file.

  Attempting to reverse engineer their code for learning purposes showed that ffmpeg is inside a compiled .so library which also seems to be where the actual connection code happens which we were unable to dig into.

  In the images 192.168.72.33 is my phone and 192.168.72.173 is the 802W device.

  Image of what I believe is a UDP broadcast of the video information.
  

  This is what the stream turns into when the device connects using the WIFI_AVIN application.
  

• swscale/aarch64 : use multiply accumulate and increase vector factor to 4

  17 novembre 2019, par Sebastian Pop

  swscale/aarch64 : use multiply accumulate and increase vector factor to 4
  This patch implements ff_hscale_8_to_15_neon with NEON fused multiply accumulate
  
  and bumps the vectorization factor from 2 to 4.
  
  The speedup is of 25% on Graviton1 A1 instances based on A-72 cpus :
  $ ffmpeg -nostats -f lavfi -i testsrc2=4k:d=2 -vf bench=start,scale=1024x1024,bench=stop -f null -
  
  before : t:0.040303 avg:0.040287 max:0.040371 min:0.039214
  
  after :  t:0.032168 avg:0.032215 max:0.033081 min:0.032146
  The speedup is of 39% on Graviton2 m6g instances based on Neoverse-N1 cpus :
  
  $ ffmpeg -nostats -f lavfi -i testsrc2=4k:d=2 -vf bench=start,scale=1024x1024,bench=stop -f null -
  
  before : t:0.019446 avg:0.019423 max:0.019493 min:0.019181
  
  after :  t:0.014015 avg:0.014096 max:0.015018 min:0.013971
  Tested with `make check` on aarch64-linux.
  Signed-off-by : Sebastian Pop <spop@amazon.com>
  
  Reviewed-by : Jean-Baptiste Kempf <jb@videolan.org>
  
  Signed-off-by : Michael Niedermayer <michael@niedermayer.cc>
  

  • [DH] libswscale/aarch64/hscale.S