Recherche avancée

Sur d’autres sites (12060)

• Revision 4d2c376923 : Early termination in encoding partition search In the partition search, the enc

  15 août 2014, par Yunqing Wang

  Changed Paths :
   Modify /vp9/encoder/vp9_context_tree.h

  
   Modify /vp9/encoder/vp9_encodeframe.c

  
   Modify /vp9/encoder/vp9_rdopt.c

  
   Modify /vp9/encoder/vp9_speed_features.c

  
   Modify /vp9/encoder/vp9_speed_features.h

  
  
  Early termination in encoding partition search

  In the partition search, the encoder checks all possible
  partitionings in the superblock’s partition search tree.
  This patch proposed a set of criteria for partition search
  early termination, which effectively decided whether or
  not to terminate the search in current branch based on the
  "skippable" result of the quantized transform coefficients.
  The "skippable" information was gathered during the
  partition mode search, and no overhead calculations were
  introduced.

  This patch gives significant encoding speed gains without
  sacrificing the quality.

  Borg test results :
  1. At speed 1,
  stdhd set : psnr : +0.074%, ssim : +0.093% ;
  derf set : psnr : -0.024%, ssim : +0.011% ;
  2. At speed 2,
  stdhd set : psnr : +0.033%, ssim : +0.100% ;
  derf set : psnr : -0.062%, ssim : +0.003% ;
  3. At speed 3,
  stdhd set : psnr : +0.060%, ssim : +0.190% ;
  derf set : psnr : -0.064%, ssim : -0.002% ;
  4. At speed 4,
  stdhd set : psnr : +0.070%, ssim : +0.143% ;
  derf set : psnr : -0.104%, ssim : +0.039% ;

  The speedup ranges from several percent to 60+%.
  speed1 speed2 speed3 speed4
  (1080p, 100f) :
  old_town_cross : 48.2% 23.9% 20.8% 16.5%
  park_joy : 11.4% 17.8% 29.4% 18.2%
  pedestrian_area : 10.7% 4.0% 4.2% 2.4%
  (720p, 200f) :
  mobcal : 68.1% 36.3% 34.4% 17.7%
  parkrun : 15.8% 24.2% 37.1% 16.8%
  shields : 45.1% 32.8% 30.1% 9.6%
  (cif, 300f)
  bus : 3.7% 10.4% 14.0% 7.9%
  deadline : 13.6% 14.8% 12.6% 10.9%
  mobile : 5.3% 11.5% 14.7% 10.7%

  Change-Id : I246c38fb952ad762ce5e365711235b605f470a66

• Packing a binary into AWS Lambda for Python

  17 novembre 2016, par Anthony G

  I want to convert m4a files uploaded to S3 to mp3. The files would only be 15 seconds max, so using Elastic Transcoder would be overkill. I downloaded a binary from https://www.johnvansickle.com/ffmpeg/. But I am very new to AWS and im still not sure how uploading binaries works. How would I would include it so that I could convert a file ?

  import boto3
  
  import urllib
  
  
  
  print('Loading function')
  
  
  
  s3 = boto3.client('s3')
  
  
  
  def lambda_handler(event, context):
  
      bucket = event['Records'][0]['s3']['bucket']['name']
  
      key = urllib.unquote_plus(event['Records'][0]['s3']['object']['key'].encode('utf8'))
  
      try:
  
      #convert to mp3
  
      #upload to bucket
  
      except Exception as e:
  
          print(e)
  
          print('Error getting object {} from bucket {}. Make sure they exist and your bucket is in the same region as this function.'.format(key, bucket))
  
          raise e

• NVIDIA accelerated ffmpeg (nvenc_h264 ) is dead slow on Amazon G2 instance

  16 novembre 2015, par Abhijit Pathak

  FFMPEG when compiled with following options give abysmal performance on Amazon G2 instance( g2.2xlarge, GRID K520) when compared with standalone Dell-Precision-T1700 (Quadro-k620)

  "--enable-nonfree --enable-gpl --enable-version3 --enable-shared --enable pthreads [b]--enable-nvenc[/b] --enable-runtime-cpudetect --disable-doc --enable-libmp3lame"

  Above ffmpeg binary took 66 seconds to transcode 5.22 mins BVE_Localize.mp4 file with following command.

  time ffmpeg -y -i BVE_Localize.mp4 -strict -2 -vcodec nvenc_h264 -b 5000k -acodec aac -ab 256k -f mpegts BVELocalize.ts ( took 1m6.990s on G2)

  When same ffmpeg command is executed on Dell-Precision-T1700 (Xeon Dual core, Quadro K620) based workstation takes 0m41.572s.

  I would like ffmpeg to perform better on Amazon G2 instance. What do you think I might be missing ? My Amazon G2 instance configuration is Ubuntu 14.04 64 bit, Cuda 7.0 , 352.55 drivers, MSI disabled,NVIDIA SDK 5.0.1