Recherche avancée

Sur d’autres sites (7048)

• dnn/native : add native support for minimum

  26 avril 2020, par Guo, Yejun

  dnn/native : add native support for minimum
  it can be tested with model file generated with below python script :
  
  import tensorflow as tf
  
  import numpy as np
  
  import imageio
  in_img = imageio.imread('input.jpg')
  
  in_img = in_img.astype(np.float32)/255.0
  
  in_data = in_img[np.newaxis, :]
  x = tf.placeholder(tf.float32, shape=[1, None, None, 3], name='dnn_in')
  
  x1 = tf.minimum(0.7, x)
  
  x2 = tf.maximum(x1, 0.4)
  
  y = tf.identity(x2, name='dnn_out')
  sess=tf.Session()
  
  sess.run(tf.global_variables_initializer())
  graph_def = tf.graph_util.convert_variables_to_constants(sess, sess.graph_def, ['dnn_out'])
  
  tf.train.write_graph(graph_def, '.', 'image_process.pb', as_text=False)
  print("image_process.pb generated, please use \
  
  path_to_ffmpeg/tools/python/convert.py to generate image_process.model\n")
  output = sess.run(y, feed_dict=x : in_data)
  
  imageio.imsave("out.jpg", np.squeeze(output))
  Signed-off-by : Guo, Yejun <yejun.guo@intel.com>
  

  • [DH] libavfilter/dnn/dnn_backend_native_layer_mathbinary.c
  • [DH] libavfilter/dnn/dnn_backend_native_layer_mathbinary.h
  • [DH] tools/python/convert_from_tensorflow.py
  • [DH] tools/python/convert_header.py

• configure : fix build issue of vf_dnn_processing.c when —disable-swscale

  2 avril 2020, par Guo, Yejun

  configure : fix build issue of vf_dnn_processing.c when —disable-swscale
  vf_dnn_processing.c recently changed to use swscale to trasfer data
  
  between AVFrame and dnn model.
  Signed-off-by : Guo, Yejun <yejun.guo@intel.com>
  
  Signed-off-by : Linjie Fu <linjie.fu@intel.com>
  

  • [DH] configure

• Android apk file too big when using FFMPeg encoder library

  9 novembre 2022, par Diego Perez

  I'm developing an app which creates x264 videos with the following library :

  



  com.arthenica:mobile-ffmpeg-full:4.2.2.LTS


  



  but the result apk file is too big ( 71mb), so I tried :

  



  com.arthenica:mobile-ffmpeg-min-gpl:4.2.2.LTS


  



  and this way, as the library downloads only a few codecs -included the ones I need- apk size was reduced to 49mb, what is much better, but still looks too big to me, so I'd need to know if any of you know a better way to reduce apk size, because people generally refuse to download such big apps

  



  Regarding the rest of the app (drawables, resources and so) they're well optimized, because if I remove this library and rebuild, the size of the app drops to 10mb

  



  I was reading this question :
FFMPEG Android Library Increase Size

  



  and user S.R suggests to compress all cpu architecture models in one archive file and extract target cpu lib based on cpu model on app's directory and load ffmpeg from there, but I really don't know how to do that.

  



  I'm checking my app's folder structure and noticed there are this next folders regarding ffmpeg lib :

  



  arm64-v8a => ~16mb
armeabi-v7a => ~29mb
x86 => ~17mb
x86_64 => ~21mb


  



  But not sure if I could remove any of them, and as you can see armeabi-v7a is the largest.