Recherche avancée

Sur d’autres sites (8273)

• ffmpeg segment naming - Moving to superuser (with apologies)

  21 juillet 2022, par kenneth558

  Maybe I am misinterpreting the downvote, but I conclude this question should go on superuser instead. Moving now and will delete this copy soon....

  


  I need to ensure unique segment names : Apparent POE cable defects, etc. around campus cause HikVision camera streams to require their ffmpeg daemons re-started once or twice or more times/day. (I am miles away from this campus for the most part, so I prefer a command line fix until the hardware fixes get applied.) When ffmpeg has to be restarted for a camera (by background bash script), I need the names of the new .mp4 segments positively not to be the same as any previous names.

  


  Background bash process currently does fine to specify an acceptable ddHHMM style new starting name for the first segment after ffmpeg restart BUT after the first or sometime second or third segment is made, ffmpeg insists on future naming to default to an unacceptable YYYmmdd style and thus start to overwrite previous segments. I use "$(date +%d%H%M)" to obtain my acceptable date style.

  


  I've tried a lot of different combinations of date codes and date embedding and both ssegment and segment muxers ; also I know very little of the very complex realm that ffmpeg is normally used in outside of simple rtsp stream copy to .mp4 files.

  


  ffmpeg command that is launched from inside bash script :
bash -c 'nohup ffmpeg -nostdin -stimeout 10000000 -rtsp_transport udp -i "rtsp://192.168.0.11:6554/Streaming/channels/101" -reconnect 1 -reconnect_at_eof 1 -reconnect_streamed 1 -c:v libx264 -f ssegment -strftime 0 -segment_time 180 -segment_format_options movflags=+faststart -reset_timestamps 0 -increment_tc 1 -avoid_negative_ts 1 -c copy -flags +global_header /var/www/camera_streams/camera_east_driveway/"$(date +%d%H%M)"_%3d.mp4 > /dev/null 2>/dev/null & '

  


  Can the segment naming pattern be carried forward indefinitely like I want ? Honestly, I wonder if ffmpeg does not allow for my specific use case naming need ?

  


  (Yes, I know changing from udp to tcp can help, but I don't consider it to be the specific solid naming fix I'm hoping for right now. And I mention HikVision in case there is known frame encoding differences for them than other cameras)

  


• qsv : Move down the implementation query

  24 février 2016, par Luca Barbato

  qsv : Move down the implementation query
  The plugin loaded may not match the general implementation capability
  
  wise.
  

  • [DBH] libavcodec/qsv.c

• dnn/vf_dnn_detect.c : add tensorflow output parse support

  6 mai 2021, par Ting Fu

  dnn/vf_dnn_detect.c : add tensorflow output parse support
  Testing model is tensorflow offical model in github repo, please refer
  
  https://github.com/tensorflow/models/blob/master/research/object_detection/g3doc/tf2_detection_zoo.md
  
  to download the detect model as you need.
  
  For example, local testing was carried on with 'ssd_mobilenet_v2_coco_2018_03_29.tar.gz', and
  
  used one image of dog in
  
  https://github.com/tensorflow/models/blob/master/research/object_detection/test_images/image1.jpg
  Testing command is :
  
  ./ffmpeg -i image1.jpg -vf dnn_detect=dnn_backend=tensorflow:input=image_tensor:output=\
  
  "num_detections&detection_scores&detection_classes&detection_boxes":model=ssd_mobilenet_v2_coco.pb,\
  
  showinfo -f null -
  We will see the result similar as below :
  
  [Parsed_showinfo_1 @ 0x33e65f0]   side data - detection bounding boxes :
  
  [Parsed_showinfo_1 @ 0x33e65f0] source : ssd_mobilenet_v2_coco.pb
  
  [Parsed_showinfo_1 @ 0x33e65f0] index : 0,       region : (382, 60) -> (1005, 593), label : 18, confidence : 9834/10000.
  
  [Parsed_showinfo_1 @ 0x33e65f0] index : 1,       region : (12, 8) -> (328, 549), label : 18, confidence : 8555/10000.
  
  [Parsed_showinfo_1 @ 0x33e65f0] index : 2,       region : (293, 7) -> (682, 458), label : 1, confidence : 8033/10000.
  
  [Parsed_showinfo_1 @ 0x33e65f0] index : 3,       region : (342, 0) -> (690, 325), label : 1, confidence : 5878/10000.
  There are two boxes of dog with cores 94.05% & 93.45% and two boxes of person with scores 80.33% & 58.78%.
  Signed-off-by : Ting Fu <ting.fu@intel.com>
  
  Signed-off-by : Guo, Yejun <yejun.guo@intel.com>
  

  • [DH] libavfilter/vf_dnn_detect.c